End-to-end learning of pharmacological assays from high-resolution microscopy images

[agitter]

We discussed this manuscript https://openreview.net/forum?id=S1gBgnR9Y7

Their setting is to use 1/3 of the images in the Cell Painting dataset that area also present in CHEMBL to predict 209 CHEMBL assay outcomes. They have several architectures to predict chemical activity using the Cell Painting images as input.

A few limitations:

• They do not compare to standard chemical features like fingerprints as a baseline
• They do not discuss or account for batch effects that we have observed

Using a semi-supervised setting could be appealing. The unlabeled images can still help constrain the representation learning, as in the Carpenter lab weak supervision paper.

[agitter]

Reproducing this paper would be difficult because we do not have their code or data. If we want to conduct a similar study and show the importance of controlling for batch effects, we could use the ExCAPE database for chemical labels. ExCAPE merges PubChem and CHEMBL and cleans the dataset: https://doi.org/10.1186%2Fs13321-017-0203-5

The next step can be to study the data they make available and see how easy it is to use and prepare a training dataset. From their paper:

Interactive access as well as download links of user selected subsets or the entire dataset are available at https://solr.ideaconsult.net/search/excape/. The whole dataset is also available at https://zenodo.org/record/173258.

We want to see what compound identifiers they provide so we can match compounds with the Cell Painting data and compute the intersection. If we can, we may want to use InChiKey as the compound identifier instead of SMILES. RDKit supports InChi: http://www.rdkit.org/Python_Docs/rdkit.Chem.inchi-module.html

[agitter]

https://github.com/xiaohk/pharmaco-image/blob/56ccbdf4a508c04e911994aee5fef797ca51b0f1/explore_excape_db.ipynb creates an initial multi-task training dataset from ExCAPE. This version uses a simpler compound matching criterion compared to converting the InChi identifiers in RDKit. If we want to expand the dataset, we could use the version with converted InChi identifiers, but I would first need to ask Scott about the RDKit warnings.

We are now set up to start training models. One baseline that the ICLR paper neglected is a fingerprint-based model. We eventually may do a hyperparameter search, but a quick assessment would be to use the best random forest from our PriA-SSB study. That was the RF_h model: The RF baseline would be single task, so we can drop all of the missing values from the column before training and evaluating. Then we do cross validation on that column, ignoring all the others. It will likely fail for columns with only 10s of non-missing rows but it could work well for columns with enough training examples.

Then we could move to simple models that operate directly on the images. RF on the images may be easy to train even if it isn't an ideal model. We will need to remember than any basic model will be biased by the batch effects.

The next intermediate could be to use the already-extracted Inception CNN model features and then train a single task model for each assay on those features. It could be RF again or logistic regression to simulate the end-to-end model we may eventually build.

We do not understand how they create their multi-task labels. It is described as a tensor where there is a length 4 dimension that has different binary labels for each of the pChEMBL thresholds (5.5, 6.5, 7.5) and the activity comment. But the training seems to be on a flattened 2d matrix instead of a tensor.

[xiaohk]

Output Matrix

0c298e084f253dcdc379a833265892c441206f1e has found 301 collisions in InChI conversion (v1 to standard). Among those, there are 296 2-collisions, 4 3-collisions, and 1 4-collision. For now, we can exclude those collisions.

Below is a summary for the output matrix. First two rows are the results after filtering compounds with at least 10 positives and 10 negatives. rdkit Conversion would have 212 assays if we don't exclude collisions.

	Compounds	Assays	Positive Proportion	Negative Proportion	NA Proportion
Naive Conversion	9404	141	1.06%	25.92%	73.02%
rdkit Conversion	26638	209	0.60%	21.29%	78.11%
rdkit Conversion	27241	212	0.61%	21.25%	78.14%
ICCR Paper	10574	209	0.87%	1.64%	97.49%

Fingerprint Random Forest

e8e735863dbac5b83bbc915e3527ccdc628f4ab5 adds the result for using fingerprint feature to train random forest single tasks models.

For each assay, we have used a stratified 5-fold cross validation, preserving positive sample proportion in each fold, to measure the classification performance on test sets. The final reported metrics are the average from 5 test sets.

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
Naive Conversion	141	10.03%	91.96%	30.35%	72.19%	27.76%	7.83%
rdkit Conversion	209	13.18%	90.72%	31.74%	71.22%	30.44%	11.43%

Here is the results from the ICCR paper:

We expect to find some correlations between metrics and pos/neg/sample size, so I added two plots:

• Accuracies are pretty high except some outliers with few assays. It is due to the skewness of positive samples.
• Our baseline model performs poorly based on metrics F1, AP, Precision, Recall.
• However, this model has a high AUC, which is the main metric used in the ICCR paper.
• Using AUC, finterprint baseline gives a similar result as ICCR paper's advanced models.
• AP has a very large variance. Low sample size (both pos and neg) tends to give high AP values.

[agitter]

Excellent analysis @xiaohk. This gives us a very strong baseline to compare to.

Are you able to sort the assays in the bottom two figures by the number of compounds (or active compounds) from smallest to largest? That would emphasize the trends even more.

[xiaohk]

rdkit gives warning Omitted undefined stereo and/or Charges were rearranged

[agitter]

We'll have to think about how to compare the performance of image-based versus fingerprint-based classification. The image-based classifier has more instances than the fingerprint-based classifier. To make them perfectly comparable, would we need to make a single aggregate prediction for all images corresponding to a unique compound and evaluate that one time instead of once per image?

[xiaohk]

c1562636ff6577cf8f5f50bd02f058a098b9637e adds results for logistic regression model with Inception 4096 features.

I agree we should control the sample size to make direct comparison with our fingerprint model. I will retrain the fingerprint model with 212 assays, and retrain an inception lr model with single random image for each compound.

Below is an initial comparison with 209 assays fingerprint random forest model:

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
FP rdkit Conversion	209	13.18%	90.72%	31.74%	71.22%	30.44%	11.43%
LR rdkit Conversion (Ignore Collision)	212-4	34.87%	84.48%	33.12%	85.22%	29.77%	75.17%

• Even though the accuracy drops, other average metrics are better than fingerprint random forest.
• Accuracy has a larger variance (matching AUC).
• This models gives 84.48% AUC, which is way better than the ICLR paper.
• Similarly to the fingerprint random forest, this model struggles with high positive size, but it is more sensitive to positive samples (high recall).

[agitter]

Is there a way to upsample the fingerprint predictions instead of downsampling the image predictions? That would entail doing something like making the same prediction for the compound k times if there are k images that correspond to the compound.

There must be some differences between our dataset and that in the ICLR paper if the AUC of our simple model is better than theirs. The only resolution may be to implement one of the complex CNNs on our version of the data.

This LR baseline implementation is done, but we could rerun it once we download and process all images with the larger quota.

[agitter]

Once we finish some validation and sanity checks in #9, we can return to this original question of whether fingerprints or image features are more informative for predicting ChEMBL bioactivity. We will still need to decide how to make the comparison as fair as possible so that both methods are predicting on the same images and the evaluation metrics are comparable.

[xiaohk]

I am running the assay activity prediction on normalized CNN features. There are 126 assays out of 212 finished. Unfortunately, it seems the performance drops after normalization.

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
LR normalized	126	24.73%	63.80%	23.41%	57.29%	22.11%	46.07%

Update (209 assays):

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
LR normalized	209	17.02%	62.64%	15.54%	56.80%	14.59%	46.73%

Some Numbers

• There are 27241 overlapping compounds, 212 assays having > 10 positives, and > 10 negatives
• Among 406 plates, we found 26939 compounds over 27241 total intersected compounds. I did not check, but I think missing 302 compounds are caused by our InChi string conversion
• 659267 images out of 913930 images (counting DMSO) are used
• I am now using the SGD logistic regression. It does not really improve the training speed significantly, and still demands a lot of memory.
• Right now, at least 35 jobs need > 32 GB RAM, 1 job needs > 100 GB RAM

The performance drop is more pronounced in the plot:

Updated plot (209 assays):

[agitter]

There is some possibility the normalization is too strong, but that is hard to directly assess because we normalize the extracted features instead of the original images.

SGD logistic regression may not be training well. Random forest may require a lot of memory to train, but we could consider it or another classifier as an alternative to see if these results are consistent across classifiers. If there are < 30,000 instances per task, random forest may be feasible. We could check a few before running all tasks.

There is also a possibility that features extracted from these images are actually not predictive of the ChEMBL activity and the poor performance is an accurate result.

[xiaohk]

Here is the result using random forest to predict drug bioactivity. I used the same random forest parameter as we were using fingerprint features.

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
RF normalized	210	8.44%	90.49%	22.11%	70.69%	22.38%	8.81%

• There is one job requiring more than 100GB RAM. For all other jobs, 100GB RAM is enough.
• RF is very slow. The run is about 0+01:04:58 to 4+12:09:33. There are many jobs running for more than one day.
• Compared to LR, the recall is extremely low, but RF has much better accuracy and AUC. Perhaps it is because RF has a stable learning experience and it learns to predict "active" for majority assays.
• The result is closed to our LR with pre-normalized features.

[xiaohk]

I think the given CellProfiler features (well average) also give a very nice baseline model. I have run a Lasso logistic regression with these features. Below is the result:

	Assays	F1	Accuracy	Average Precision	AUC	Precision	Recall
Mean well feature	206	24.74%	81.07%	24.54%	69.52%	20.83%	42.27%.

• The feature matrix has size (110622, 1786) with 26939 overlapping compounds

• There are 1800 given features. I didn't use any categorical features:

  ['Metadata_Well', 'Metadata_Plate_Map_Name',

  'Metadata_well_position', 'Metadata_ASSAY_WELL_ROLE', 'Metadata_broad_sample', 'Metadata_solvent', 'Metadata_pert_id', 'Metadata_pert_mfc_id', 'Metadata_pert_well', 'Metadata_cell_id', 'Metadata_broad_sample_type', 'Metadata_pert_vehicle', 'Metadata_pert_type']

• This model makes a better comparison to our fingerprint models. It does not use feature at image level, but at well level (because we are averaging the cell profiler features).
• I am not sure if they have normalized these features
• It is very slow. 26 jobs were put in hold after running for 3 days
• The result looks like our fingerprint model

[agitter]

Running CellProfiler features gives some hope there is still some signal in the image data. Averaging over the cells could potentially be boosting the performance by making the predictions less sensitive to the number of cells in the image. The highly-variable number of cells could lead to very different extracted feature representations even for cells that are biologically similar. We can look into the variance of cell count in one well.

The Random Forest is performing better, but average precision is still quite low for many of the assays that have larger sample sizes are are highly skewed negative in class label.

We could try standard logistic regression instead of stochastic.

Did the ICLR submission have implementations available for the methods that operate in individual cropped cells? Perhaps we could look into what models we would want to train if we have the ability to do so (#11).

[agitter]

@xiaohk had the good idea of selecting some assays for which there are enough examples to train models but not too many to make model training and data transfer unreasonably slow.

[agitter]

In #12 they also only subselect assays that have good AUC-ROC. This suggests that we do not need to show that image-based features work for all assays, only that some assays can be predicted well given the images. We can focus on finding assays with a reasonable number of tested compounds and reasonable performance with our initial-image based models, perhaps the traditional models using Cell Profiler features.

We are also wondering whether we can access the median of the Cell Profiler features by using a different database table.

[xiaohk]

e026377229351f9a8a812ae530bf49cf033afa92 adds routines to select assays and extract median CellProfiler features.

Median CellProfiler Features

For each plate, we can access to one SQL table at the image level Images, and three SQL tables at the cell level Cells, Cytoplasm and Nuclei.

For each image (per field of view), I combine its Images features along with the median of Cells, Cytoplasm and Nuclei features. String features and database metadata features are not extracted. Therefore, we have a 2366 length vector to represent aggregated CellProfiler features for each image.

Image	Cells	Cytoplasm	Nuclei
Number of features	583	596	582	605

All selected features are listed below:

Image Table

'Image_Count_Cells', 'Image_Count_Cytoplasm', 'Image_Count_Nuclei', 'Image_ExecutionTime_01LoadData', 'Image_ExecutionTime_02CorrectIlluminationApply', 'Image_ExecutionTime_03MeasureImageQuality', 'Image_ExecutionTime_04MeasureImageQuality', 'Image_ExecutionTime_06IdentifyPrimaryObjects', 'Image_ExecutionTime_07IdentifySecondaryObjects', 'Image_ExecutionTime_08IdentifyTertiaryObjects', 'Image_ExecutionTime_09MeasureCorrelation', 'Image_ExecutionTime_10MeasureGranularity', 'Image_ExecutionTime_11MeasureObjectIntensity', 'Image_ExecutionTime_12MeasureObjectNeighbors', 'Image_ExecutionTime_13MeasureObjectNeighbors', 'Image_ExecutionTime_14MeasureObjectNeighbors', 'Image_ExecutionTime_15MeasureObjectIntensityDistribution', 'Image_ExecutionTime_16MeasureObjectSizeShape', 'Image_ExecutionTime_17MeasureTexture', 'Image_ExecutionTime_18OverlayOutlines', 'Image_ExecutionTime_19OverlayOutlines', 'Image_ExecutionTime_20SaveImages', 'Image_ExecutionTime_21SaveImages', 'Image_Granularity_10_AGP', 'Image_Granularity_10_DNA', 'Image_Granularity_10_ER', 'Image_Granularity_10_Mito', 'Image_Granularity_10_RNA', 'Image_Granularity_11_AGP', 'Image_Granularity_11_DNA', 'Image_Granularity_11_ER', 'Image_Granularity_11_Mito', 'Image_Granularity_11_RNA', 'Image_Granularity_12_AGP', 'Image_Granularity_12_DNA', 'Image_Granularity_12_ER', 'Image_Granularity_12_Mito', 'Image_Granularity_12_RNA', 'Image_Granularity_13_AGP', 'Image_Granularity_13_DNA', 'Image_Granularity_13_ER', 'Image_Granularity_13_Mito', 'Image_Granularity_13_RNA', 'Image_Granularity_14_AGP', 'Image_Granularity_14_DNA', 'Image_Granularity_14_ER', 'Image_Granularity_14_Mito', 'Image_Granularity_14_RNA', 'Image_Granularity_15_AGP', 'Image_Granularity_15_DNA', 'Image_Granularity_15_ER', 'Image_Granularity_15_Mito', 'Image_Granularity_15_RNA', 'Image_Granularity_16_AGP', 'Image_Granularity_16_DNA', 'Image_Granularity_16_ER', 'Image_Granularity_16_Mito', 'Image_Granularity_16_RNA', 'Image_Granularity_1_AGP', 'Image_Granularity_1_DNA', 'Image_Granularity_1_ER', 'Image_Granularity_1_Mito', 'Image_Granularity_1_RNA', 'Image_Granularity_2_AGP', 'Image_Granularity_2_DNA', 'Image_Granularity_2_ER', 'Image_Granularity_2_Mito', 'Image_Granularity_2_RNA', 'Image_Granularity_3_AGP', 'Image_Granularity_3_DNA', 'Image_Granularity_3_ER', 'Image_Granularity_3_Mito', 'Image_Granularity_3_RNA', 'Image_Granularity_4_AGP', 'Image_Granularity_4_DNA', 'Image_Granularity_4_ER', 'Image_Granularity_4_Mito', 'Image_Granularity_4_RNA', 'Image_Granularity_5_AGP', 'Image_Granularity_5_DNA', 'Image_Granularity_5_ER', 'Image_Granularity_5_Mito', 'Image_Granularity_5_RNA', 'Image_Granularity_6_AGP', 'Image_Granularity_6_DNA', 'Image_Granularity_6_ER', 'Image_Granularity_6_Mito', 'Image_Granularity_6_RNA', 'Image_Granularity_7_AGP', 'Image_Granularity_7_DNA', 'Image_Granularity_7_ER', 'Image_Granularity_7_Mito', 'Image_Granularity_7_RNA', 'Image_Granularity_8_AGP', 'Image_Granularity_8_DNA', 'Image_Granularity_8_ER', 'Image_Granularity_8_Mito', 'Image_Granularity_8_RNA', 'Image_Granularity_9_AGP', 'Image_Granularity_9_DNA', 'Image_Granularity_9_ER', 'Image_Granularity_9_Mito', 'Image_Granularity_9_RNA', 'Image_Height_IllumAGP', 'Image_Height_IllumDNA', 'Image_Height_IllumER', 'Image_Height_IllumMito', 'Image_Height_IllumRNA', 'Image_Height_OrigAGP', 'Image_Height_OrigDNA', 'Image_Height_OrigER', 'Image_Height_OrigMito', 'Image_Height_OrigRNA', 'Image_ImageQuality_Correlation_IllumAGP_10', 'Image_ImageQuality_Correlation_IllumAGP_2', 'Image_ImageQuality_Correlation_IllumAGP_20', 'Image_ImageQuality_Correlation_IllumAGP_5', 'Image_ImageQuality_Correlation_IllumDNA_10', 'Image_ImageQuality_Correlation_IllumDNA_2', 'Image_ImageQuality_Correlation_IllumDNA_20', 'Image_ImageQuality_Correlation_IllumDNA_5', 'Image_ImageQuality_Correlation_IllumER_10', 'Image_ImageQuality_Correlation_IllumER_2', 'Image_ImageQuality_Correlation_IllumER_20', 'Image_ImageQuality_Correlation_IllumER_5', 'Image_ImageQuality_Correlation_IllumMito_10', 'Image_ImageQuality_Correlation_IllumMito_2', 'Image_ImageQuality_Correlation_IllumMito_20', 'Image_ImageQuality_Correlation_IllumMito_5', 'Image_ImageQuality_Correlation_IllumRNA_10', 'Image_ImageQuality_Correlation_IllumRNA_2', 'Image_ImageQuality_Correlation_IllumRNA_20', 'Image_ImageQuality_Correlation_IllumRNA_5', 'Image_ImageQuality_Correlation_OrigAGP_10', 'Image_ImageQuality_Correlation_OrigAGP_2', 'Image_ImageQuality_Correlation_OrigAGP_20', 'Image_ImageQuality_Correlation_OrigAGP_5', 'Image_ImageQuality_Correlation_OrigDNA_10', 'Image_ImageQuality_Correlation_OrigDNA_2', 'Image_ImageQuality_Correlation_OrigDNA_20', 'Image_ImageQuality_Correlation_OrigDNA_5', 'Image_ImageQuality_Correlation_OrigER_10', 'Image_ImageQuality_Correlation_OrigER_2', 'Image_ImageQuality_Correlation_OrigER_20', 'Image_ImageQuality_Correlation_OrigER_5', 'Image_ImageQuality_Correlation_OrigMito_10', 'Image_ImageQuality_Correlation_OrigMito_2', 'Image_ImageQuality_Correlation_OrigMito_20', 'Image_ImageQuality_Correlation_OrigMito_5', 'Image_ImageQuality_Correlation_OrigRNA_10', 'Image_ImageQuality_Correlation_OrigRNA_2', 'Image_ImageQuality_Correlation_OrigRNA_20', 'Image_ImageQuality_Correlation_OrigRNA_5', 'Image_ImageQuality_FocusScore_IllumAGP', 'Image_ImageQuality_FocusScore_IllumDNA', 'Image_ImageQuality_FocusScore_IllumER', 'Image_ImageQuality_FocusScore_IllumMito', 'Image_ImageQuality_FocusScore_IllumRNA', 'Image_ImageQuality_FocusScore_OrigAGP', 'Image_ImageQuality_FocusScore_OrigDNA', 'Image_ImageQuality_FocusScore_OrigER', 'Image_ImageQuality_FocusScore_OrigMito', 'Image_ImageQuality_FocusScore_OrigRNA', 'Image_ImageQuality_LocalFocusScore_IllumAGP_10', 'Image_ImageQuality_LocalFocusScore_IllumAGP_2', 'Image_ImageQuality_LocalFocusScore_IllumAGP_20', 'Image_ImageQuality_LocalFocusScore_IllumAGP_5', 'Image_ImageQuality_LocalFocusScore_IllumDNA_10', 'Image_ImageQuality_LocalFocusScore_IllumDNA_2', 'Image_ImageQuality_LocalFocusScore_IllumDNA_20', 'Image_ImageQuality_LocalFocusScore_IllumDNA_5', 'Image_ImageQuality_LocalFocusScore_IllumER_10', 'Image_ImageQuality_LocalFocusScore_IllumER_2', 'Image_ImageQuality_LocalFocusScore_IllumER_20', 'Image_ImageQuality_LocalFocusScore_IllumER_5', 'Image_ImageQuality_LocalFocusScore_IllumMito_10', 'Image_ImageQuality_LocalFocusScore_IllumMito_2', 'Image_ImageQuality_LocalFocusScore_IllumMito_20', 'Image_ImageQuality_LocalFocusScore_IllumMito_5', 'Image_ImageQuality_LocalFocusScore_IllumRNA_10', 'Image_ImageQuality_LocalFocusScore_IllumRNA_2', 'Image_ImageQuality_LocalFocusScore_IllumRNA_20', 'Image_ImageQuality_LocalFocusScore_IllumRNA_5', 'Image_ImageQuality_LocalFocusScore_OrigAGP_10', 'Image_ImageQuality_LocalFocusScore_OrigAGP_2', 'Image_ImageQuality_LocalFocusScore_OrigAGP_20', 'Image_ImageQuality_LocalFocusScore_OrigAGP_5', 'Image_ImageQuality_LocalFocusScore_OrigDNA_10', 'Image_ImageQuality_LocalFocusScore_OrigDNA_2', 'Image_ImageQuality_LocalFocusScore_OrigDNA_20', 'Image_ImageQuality_LocalFocusScore_OrigDNA_5', 'Image_ImageQuality_LocalFocusScore_OrigER_10', 'Image_ImageQuality_LocalFocusScore_OrigER_2', 'Image_ImageQuality_LocalFocusScore_OrigER_20', 'Image_ImageQuality_LocalFocusScore_OrigER_5', 'Image_ImageQuality_LocalFocusScore_OrigMito_10', 'Image_ImageQuality_LocalFocusScore_OrigMito_2', 'Image_ImageQuality_LocalFocusScore_OrigMito_20', 'Image_ImageQuality_LocalFocusScore_OrigMito_5', 'Image_ImageQuality_LocalFocusScore_OrigRNA_10', 'Image_ImageQuality_LocalFocusScore_OrigRNA_2', 'Image_ImageQuality_LocalFocusScore_OrigRNA_20', 'Image_ImageQuality_LocalFocusScore_OrigRNA_5', 'Image_ImageQuality_MADIntensity_IllumAGP', 'Image_ImageQuality_MADIntensity_IllumDNA', 'Image_ImageQuality_MADIntensity_IllumER', 'Image_ImageQuality_MADIntensity_IllumMito', 'Image_ImageQuality_MADIntensity_IllumRNA', 'Image_ImageQuality_MADIntensity_OrigAGP', 'Image_ImageQuality_MADIntensity_OrigDNA', 'Image_ImageQuality_MADIntensity_OrigER', 'Image_ImageQuality_MADIntensity_OrigMito', 'Image_ImageQuality_MADIntensity_OrigRNA', 'Image_ImageQuality_MaxIntensity_IllumAGP', 'Image_ImageQuality_MaxIntensity_IllumDNA', 'Image_ImageQuality_MaxIntensity_IllumER', 'Image_ImageQuality_MaxIntensity_IllumMito', 'Image_ImageQuality_MaxIntensity_IllumRNA', 'Image_ImageQuality_MaxIntensity_OrigAGP', 'Image_ImageQuality_MaxIntensity_OrigDNA', 'Image_ImageQuality_MaxIntensity_OrigER', 'Image_ImageQuality_MaxIntensity_OrigMito', 'Image_ImageQuality_MaxIntensity_OrigRNA', 'Image_ImageQuality_MeanIntensity_IllumAGP', 'Image_ImageQuality_MeanIntensity_IllumDNA', 'Image_ImageQuality_MeanIntensity_IllumER', 'Image_ImageQuality_MeanIntensity_IllumMito', 'Image_ImageQuality_MeanIntensity_IllumRNA', 'Image_ImageQuality_MeanIntensity_OrigAGP', 'Image_ImageQuality_MeanIntensity_OrigDNA', 'Image_ImageQuality_MeanIntensity_OrigER', 'Image_ImageQuality_MeanIntensity_OrigMito', 'Image_ImageQuality_MeanIntensity_OrigRNA', 'Image_ImageQuality_MedianIntensity_IllumAGP', 'Image_ImageQuality_MedianIntensity_IllumDNA', 'Image_ImageQuality_MedianIntensity_IllumER', 'Image_ImageQuality_MedianIntensity_IllumMito', 'Image_ImageQuality_MedianIntensity_IllumRNA', 'Image_ImageQuality_MedianIntensity_OrigAGP', 'Image_ImageQuality_MedianIntensity_OrigDNA', 'Image_ImageQuality_MedianIntensity_OrigER', 'Image_ImageQuality_MedianIntensity_OrigMito', 'Image_ImageQuality_MedianIntensity_OrigRNA', 'Image_ImageQuality_MinIntensity_IllumAGP', 'Image_ImageQuality_MinIntensity_IllumDNA', 'Image_ImageQuality_MinIntensity_IllumER', 'Image_ImageQuality_MinIntensity_IllumMito', 'Image_ImageQuality_MinIntensity_IllumRNA', 'Image_ImageQuality_MinIntensity_OrigAGP', 'Image_ImageQuality_MinIntensity_OrigDNA', 'Image_ImageQuality_MinIntensity_OrigER', 'Image_ImageQuality_MinIntensity_OrigMito', 'Image_ImageQuality_MinIntensity_OrigRNA', 'Image_ImageQuality_PercentMaximal_IllumAGP', 'Image_ImageQuality_PercentMaximal_IllumDNA', 'Image_ImageQuality_PercentMaximal_IllumER', 'Image_ImageQuality_PercentMaximal_IllumMito', 'Image_ImageQuality_PercentMaximal_IllumRNA', 'Image_ImageQuality_PercentMaximal_OrigAGP', 'Image_ImageQuality_PercentMaximal_OrigDNA', 'Image_ImageQuality_PercentMaximal_OrigER', 'Image_ImageQuality_PercentMaximal_OrigMito', 'Image_ImageQuality_PercentMaximal_OrigRNA', 'Image_ImageQuality_PercentMinimal_IllumAGP', 'Image_ImageQuality_PercentMinimal_IllumDNA', 'Image_ImageQuality_PercentMinimal_IllumER', 'Image_ImageQuality_PercentMinimal_IllumMito', 'Image_ImageQuality_PercentMinimal_IllumRNA', 'Image_ImageQuality_PercentMinimal_OrigAGP', 'Image_ImageQuality_PercentMinimal_OrigDNA', 'Image_ImageQuality_PercentMinimal_OrigER', 'Image_ImageQuality_PercentMinimal_OrigMito', 'Image_ImageQuality_PercentMinimal_OrigRNA', 'Image_ImageQuality_PowerLogLogSlope_IllumAGP', 'Image_ImageQuality_PowerLogLogSlope_IllumDNA', 'Image_ImageQuality_PowerLogLogSlope_IllumER', 'Image_ImageQuality_PowerLogLogSlope_IllumMito', 'Image_ImageQuality_PowerLogLogSlope_IllumRNA', 'Image_ImageQuality_PowerLogLogSlope_OrigAGP', 'Image_ImageQuality_PowerLogLogSlope_OrigDNA', 'Image_ImageQuality_PowerLogLogSlope_OrigER', 'Image_ImageQuality_PowerLogLogSlope_OrigMito', 'Image_ImageQuality_PowerLogLogSlope_OrigRNA', 'Image_ImageQuality_Scaling_IllumAGP', 'Image_ImageQuality_Scaling_IllumDNA', 'Image_ImageQuality_Scaling_IllumER', 'Image_ImageQuality_Scaling_IllumMito', 'Image_ImageQuality_Scaling_IllumRNA', 'Image_ImageQuality_Scaling_OrigAGP', 'Image_ImageQuality_Scaling_OrigDNA', 'Image_ImageQuality_Scaling_OrigER', 'Image_ImageQuality_Scaling_OrigMito', 'Image_ImageQuality_Scaling_OrigRNA', 'Image_ImageQuality_StdIntensity_IllumAGP', 'Image_ImageQuality_StdIntensity_IllumDNA', 'Image_ImageQuality_StdIntensity_IllumER', 'Image_ImageQuality_StdIntensity_IllumMito', 'Image_ImageQuality_StdIntensity_IllumRNA', 'Image_ImageQuality_StdIntensity_OrigAGP', 'Image_ImageQuality_StdIntensity_OrigDNA', 'Image_ImageQuality_StdIntensity_OrigER', 'Image_ImageQuality_StdIntensity_OrigMito', 'Image_ImageQuality_StdIntensity_OrigRNA', 'Image_ImageQuality_ThresholdOtsu_OrigDNA_2W', 'Image_ImageQuality_ThresholdOtsu_OrigRNA_3FW', 'Image_ImageQuality_TotalArea_IllumAGP', 'Image_ImageQuality_TotalArea_IllumDNA', 'Image_ImageQuality_TotalArea_IllumER', 'Image_ImageQuality_TotalArea_IllumMito', 'Image_ImageQuality_TotalArea_IllumRNA', 'Image_ImageQuality_TotalArea_OrigAGP', 'Image_ImageQuality_TotalArea_OrigDNA', 'Image_ImageQuality_TotalArea_OrigER', 'Image_ImageQuality_TotalArea_OrigMito', 'Image_ImageQuality_TotalArea_OrigRNA', 'Image_ImageQuality_TotalIntensity_IllumAGP', 'Image_ImageQuality_TotalIntensity_IllumDNA', 'Image_ImageQuality_TotalIntensity_IllumER', 'Image_ImageQuality_TotalIntensity_IllumMito', 'Image_ImageQuality_TotalIntensity_IllumRNA', 'Image_ImageQuality_TotalIntensity_OrigAGP', 'Image_ImageQuality_TotalIntensity_OrigDNA', 'Image_ImageQuality_TotalIntensity_OrigER', 'Image_ImageQuality_TotalIntensity_OrigMito', 'Image_ImageQuality_TotalIntensity_OrigRNA', 'Image_ModuleError_01LoadData', 'Image_ModuleError_02CorrectIlluminationApply', 'Image_ModuleError_03MeasureImageQuality', 'Image_ModuleError_04MeasureImageQuality', 'Image_ModuleError_06IdentifyPrimaryObjects', 'Image_ModuleError_07IdentifySecondaryObjects', 'Image_ModuleError_08IdentifyTertiaryObjects', 'Image_ModuleError_09MeasureCorrelation', 'Image_ModuleError_10MeasureGranularity', 'Image_ModuleError_11MeasureObjectIntensity', 'Image_ModuleError_12MeasureObjectNeighbors', 'Image_ModuleError_13MeasureObjectNeighbors', 'Image_ModuleError_14MeasureObjectNeighbors', 'Image_ModuleError_15MeasureObjectIntensityDistribution', 'Image_ModuleError_16MeasureObjectSizeShape', 'Image_ModuleError_17MeasureTexture', 'Image_ModuleError_18OverlayOutlines', 'Image_ModuleError_19OverlayOutlines', 'Image_ModuleError_20SaveImages', 'Image_ModuleError_21SaveImages', 'Image_Scaling_IllumAGP', 'Image_Scaling_IllumDNA', 'Image_Scaling_IllumER', 'Image_Scaling_IllumMito', 'Image_Scaling_IllumRNA', 'Image_Scaling_OrigAGP', 'Image_Scaling_OrigDNA', 'Image_Scaling_OrigER', 'Image_Scaling_OrigMito', 'Image_Scaling_OrigRNA', 'Image_Texture_AngularSecondMoment_AGP_10_0', 'Image_Texture_AngularSecondMoment_AGP_3_0', 'Image_Texture_AngularSecondMoment_AGP_5_0', 'Image_Texture_AngularSecondMoment_DNA_10_0', 'Image_Texture_AngularSecondMoment_DNA_3_0', 'Image_Texture_AngularSecondMoment_DNA_5_0', 'Image_Texture_AngularSecondMoment_ER_10_0', 'Image_Texture_AngularSecondMoment_ER_3_0', 'Image_Texture_AngularSecondMoment_ER_5_0', 'Image_Texture_AngularSecondMoment_Mito_10_0', 'Image_Texture_AngularSecondMoment_Mito_3_0', 'Image_Texture_AngularSecondMoment_Mito_5_0', 'Image_Texture_AngularSecondMoment_RNA_10_0', 'Image_Texture_AngularSecondMoment_RNA_3_0', 'Image_Texture_AngularSecondMoment_RNA_5_0', 'Image_Texture_Contrast_AGP_10_0', 'Image_Texture_Contrast_AGP_3_0', 'Image_Texture_Contrast_AGP_5_0', 'Image_Texture_Contrast_DNA_10_0', 'Image_Texture_Contrast_DNA_3_0', 'Image_Texture_Contrast_DNA_5_0', 'Image_Texture_Contrast_ER_10_0', 'Image_Texture_Contrast_ER_3_0', 'Image_Texture_Contrast_ER_5_0', 'Image_Texture_Contrast_Mito_10_0', 'Image_Texture_Contrast_Mito_3_0', 'Image_Texture_Contrast_Mito_5_0', 'Image_Texture_Contrast_RNA_10_0', 'Image_Texture_Contrast_RNA_3_0', 'Image_Texture_Contrast_RNA_5_0', 'Image_Texture_Correlation_AGP_10_0', 'Image_Texture_Correlation_AGP_3_0', 'Image_Texture_Correlation_AGP_5_0', 'Image_Texture_Correlation_DNA_10_0', 'Image_Texture_Correlation_DNA_3_0', 'Image_Texture_Correlation_DNA_5_0', 'Image_Texture_Correlation_ER_10_0', 'Image_Texture_Correlation_ER_3_0', 'Image_Texture_Correlation_ER_5_0', 'Image_Texture_Correlation_Mito_10_0', 'Image_Texture_Correlation_Mito_3_0', 'Image_Texture_Correlation_Mito_5_0', 'Image_Texture_Correlation_RNA_10_0', 'Image_Texture_Correlation_RNA_3_0', 'Image_Texture_Correlation_RNA_5_0', 'Image_Texture_DifferenceEntropy_AGP_10_0', 'Image_Texture_DifferenceEntropy_AGP_3_0', 'Image_Texture_DifferenceEntropy_AGP_5_0', 'Image_Texture_DifferenceEntropy_DNA_10_0', 'Image_Texture_DifferenceEntropy_DNA_3_0', 'Image_Texture_DifferenceEntropy_DNA_5_0', 'Image_Texture_DifferenceEntropy_ER_10_0', 'Image_Texture_DifferenceEntropy_ER_3_0', 'Image_Texture_DifferenceEntropy_ER_5_0', 'Image_Texture_DifferenceEntropy_Mito_10_0', 'Image_Texture_DifferenceEntropy_Mito_3_0', 'Image_Texture_DifferenceEntropy_Mito_5_0', 'Image_Texture_DifferenceEntropy_RNA_10_0', 'Image_Texture_DifferenceEntropy_RNA_3_0', 'Image_Texture_DifferenceEntropy_RNA_5_0', 'Image_Texture_DifferenceVariance_AGP_10_0', 'Image_Texture_DifferenceVariance_AGP_3_0', 'Image_Texture_DifferenceVariance_AGP_5_0', 'Image_Texture_DifferenceVariance_DNA_10_0', 'Image_Texture_DifferenceVariance_DNA_3_0', 'Image_Texture_DifferenceVariance_DNA_5_0', 'Image_Texture_DifferenceVariance_ER_10_0', 'Image_Texture_DifferenceVariance_ER_3_0', 'Image_Texture_DifferenceVariance_ER_5_0', 'Image_Texture_DifferenceVariance_Mito_10_0', 'Image_Texture_DifferenceVariance_Mito_3_0', 'Image_Texture_DifferenceVariance_Mito_5_0', 'Image_Texture_DifferenceVariance_RNA_10_0', 'Image_Texture_DifferenceVariance_RNA_3_0', 'Image_Texture_DifferenceVariance_RNA_5_0', 'Image_Texture_Entropy_AGP_10_0', 'Image_Texture_Entropy_AGP_3_0', 'Image_Texture_Entropy_AGP_5_0', 'Image_Texture_Entropy_DNA_10_0', 'Image_Texture_Entropy_DNA_3_0', 'Image_Texture_Entropy_DNA_5_0', 'Image_Texture_Entropy_ER_10_0', 'Image_Texture_Entropy_ER_3_0', 'Image_Texture_Entropy_ER_5_0', 'Image_Texture_Entropy_Mito_10_0', 'Image_Texture_Entropy_Mito_3_0', 'Image_Texture_Entropy_Mito_5_0', 'Image_Texture_Entropy_RNA_10_0', 'Image_Texture_Entropy_RNA_3_0', 'Image_Texture_Entropy_RNA_5_0', 'Image_Texture_Gabor_AGP_10', 'Image_Texture_Gabor_AGP_3', 'Image_Texture_Gabor_AGP_5', 'Image_Texture_Gabor_DNA_10', 'Image_Texture_Gabor_DNA_3', 'Image_Texture_Gabor_DNA_5', 'Image_Texture_Gabor_ER_10', 'Image_Texture_Gabor_ER_3', 'Image_Texture_Gabor_ER_5', 'Image_Texture_Gabor_Mito_10', 'Image_Texture_Gabor_Mito_3', 'Image_Texture_Gabor_Mito_5', 'Image_Texture_Gabor_RNA_10', 'Image_Texture_Gabor_RNA_3', 'Image_Texture_Gabor_RNA_5', 'Image_Texture_InfoMeas1_AGP_10_0', 'Image_Texture_InfoMeas1_AGP_3_0', 'Image_Texture_InfoMeas1_AGP_5_0', 'Image_Texture_InfoMeas1_DNA_10_0', 'Image_Texture_InfoMeas1_DNA_3_0', 'Image_Texture_InfoMeas1_DNA_5_0', 'Image_Texture_InfoMeas1_ER_10_0', 'Image_Texture_InfoMeas1_ER_3_0', 'Image_Texture_InfoMeas1_ER_5_0', 'Image_Texture_InfoMeas1_Mito_10_0', 'Image_Texture_InfoMeas1_Mito_3_0', 'Image_Texture_InfoMeas1_Mito_5_0', 'Image_Texture_InfoMeas1_RNA_10_0', 'Image_Texture_InfoMeas1_RNA_3_0', 'Image_Texture_InfoMeas1_RNA_5_0', 'Image_Texture_InfoMeas2_AGP_10_0', 'Image_Texture_InfoMeas2_AGP_3_0', 'Image_Texture_InfoMeas2_AGP_5_0', 'Image_Texture_InfoMeas2_DNA_10_0', 'Image_Texture_InfoMeas2_DNA_3_0', 'Image_Texture_InfoMeas2_DNA_5_0', 'Image_Texture_InfoMeas2_ER_10_0', 'Image_Texture_InfoMeas2_ER_3_0', 'Image_Texture_InfoMeas2_ER_5_0', 'Image_Texture_InfoMeas2_Mito_10_0', 'Image_Texture_InfoMeas2_Mito_3_0', 'Image_Texture_InfoMeas2_Mito_5_0', 'Image_Texture_InfoMeas2_RNA_10_0', 'Image_Texture_InfoMeas2_RNA_3_0', 'Image_Texture_InfoMeas2_RNA_5_0', 'Image_Texture_InverseDifferenceMoment_AGP_10_0', 'Image_Texture_InverseDifferenceMoment_AGP_3_0', 'Image_Texture_InverseDifferenceMoment_AGP_5_0', 'Image_Texture_InverseDifferenceMoment_DNA_10_0', 'Image_Texture_InverseDifferenceMoment_DNA_3_0', 'Image_Texture_InverseDifferenceMoment_DNA_5_0', 'Image_Texture_InverseDifferenceMoment_ER_10_0', 'Image_Texture_InverseDifferenceMoment_ER_3_0', 'Image_Texture_InverseDifferenceMoment_ER_5_0', 'Image_Texture_InverseDifferenceMoment_Mito_10_0', 'Image_Texture_InverseDifferenceMoment_Mito_3_0', 'Image_Texture_InverseDifferenceMoment_Mito_5_0', 'Image_Texture_InverseDifferenceMoment_RNA_10_0', 'Image_Texture_InverseDifferenceMoment_RNA_3_0', 'Image_Texture_InverseDifferenceMoment_RNA_5_0', 'Image_Texture_SumAverage_AGP_10_0', 'Image_Texture_SumAverage_AGP_3_0', 'Image_Texture_SumAverage_AGP_5_0', 'Image_Texture_SumAverage_DNA_10_0', 'Image_Texture_SumAverage_DNA_3_0', 'Image_Texture_SumAverage_DNA_5_0', 'Image_Texture_SumAverage_ER_10_0', 'Image_Texture_SumAverage_ER_3_0', 'Image_Texture_SumAverage_ER_5_0', 'Image_Texture_SumAverage_Mito_10_0', 'Image_Texture_SumAverage_Mito_3_0', 'Image_Texture_SumAverage_Mito_5_0', 'Image_Texture_SumAverage_RNA_10_0', 'Image_Texture_SumAverage_RNA_3_0', 'Image_Texture_SumAverage_RNA_5_0', 'Image_Texture_SumEntropy_AGP_10_0', 'Image_Texture_SumEntropy_AGP_3_0', 'Image_Texture_SumEntropy_AGP_5_0', 'Image_Texture_SumEntropy_DNA_10_0', 'Image_Texture_SumEntropy_DNA_3_0', 'Image_Texture_SumEntropy_DNA_5_0', 'Image_Texture_SumEntropy_ER_10_0', 'Image_Texture_SumEntropy_ER_3_0', 'Image_Texture_SumEntropy_ER_5_0', 'Image_Texture_SumEntropy_Mito_10_0', 'Image_Texture_SumEntropy_Mito_3_0', 'Image_Texture_SumEntropy_Mito_5_0', 'Image_Texture_SumEntropy_RNA_10_0', 'Image_Texture_SumEntropy_RNA_3_0', 'Image_Texture_SumEntropy_RNA_5_0', 'Image_Texture_SumVariance_AGP_10_0', 'Image_Texture_SumVariance_AGP_3_0', 'Image_Texture_SumVariance_AGP_5_0', 'Image_Texture_SumVariance_DNA_10_0', 'Image_Texture_SumVariance_DNA_3_0', 'Image_Texture_SumVariance_DNA_5_0', 'Image_Texture_SumVariance_ER_10_0', 'Image_Texture_SumVariance_ER_3_0', 'Image_Texture_SumVariance_ER_5_0', 'Image_Texture_SumVariance_Mito_10_0', 'Image_Texture_SumVariance_Mito_3_0', 'Image_Texture_SumVariance_Mito_5_0', 'Image_Texture_SumVariance_RNA_10_0', 'Image_Texture_SumVariance_RNA_3_0', 'Image_Texture_SumVariance_RNA_5_0', 'Image_Texture_Variance_AGP_10_0', 'Image_Texture_Variance_AGP_3_0', 'Image_Texture_Variance_AGP_5_0', 'Image_Texture_Variance_DNA_10_0', 'Image_Texture_Variance_DNA_3_0', 'Image_Texture_Variance_DNA_5_0', 'Image_Texture_Variance_ER_10_0', 'Image_Texture_Variance_ER_3_0', 'Image_Texture_Variance_ER_5_0', 'Image_Texture_Variance_Mito_10_0', 'Image_Texture_Variance_Mito_3_0', 'Image_Texture_Variance_Mito_5_0', 'Image_Texture_Variance_RNA_10_0', 'Image_Texture_Variance_RNA_3_0', 'Image_Texture_Variance_RNA_5_0', 'Image_Threshold_FinalThreshold_Cells', 'Image_Threshold_FinalThreshold_Nuclei', 'Image_Threshold_OrigThreshold_Cells', 'Image_Threshold_OrigThreshold_Nuclei', 'Image_Threshold_SumOfEntropies_Cells', 'Image_Threshold_SumOfEntropies_Nuclei', 'Image_Threshold_WeightedVariance_Cells', 'Image_Threshold_WeightedVariance_Nuclei', 'Image_Width_IllumAGP', 'Image_Width_IllumDNA', 'Image_Width_IllumER', 'Image_Width_IllumMito', 'Image_Width_IllumRNA', 'Image_Width_OrigAGP', 'Image_Width_OrigDNA', 'Image_Width_OrigER', 'Image_Width_OrigMito', 'Image_Width_OrigRNA'

Cell Table

'Cells_AreaShape_Area', 'Cells_AreaShape_Center_X', 'Cells_AreaShape_Center_Y', 'Cells_AreaShape_Compactness', 'Cells_AreaShape_Eccentricity', 'Cells_AreaShape_EulerNumber', 'Cells_AreaShape_Extent', 'Cells_AreaShape_FormFactor', 'Cells_AreaShape_MajorAxisLength', 'Cells_AreaShape_MaxFeretDiameter', 'Cells_AreaShape_MaximumRadius', 'Cells_AreaShape_MeanRadius', 'Cells_AreaShape_MedianRadius', 'Cells_AreaShape_MinFeretDiameter', 'Cells_AreaShape_MinorAxisLength', 'Cells_AreaShape_Orientation', 'Cells_AreaShape_Perimeter', 'Cells_AreaShape_Solidity', 'Cells_AreaShape_Zernike_0_0', 'Cells_AreaShape_Zernike_1_1', 'Cells_AreaShape_Zernike_2_0', 'Cells_AreaShape_Zernike_2_2', 'Cells_AreaShape_Zernike_3_1', 'Cells_AreaShape_Zernike_3_3', 'Cells_AreaShape_Zernike_4_0', 'Cells_AreaShape_Zernike_4_2', 'Cells_AreaShape_Zernike_4_4', 'Cells_AreaShape_Zernike_5_1', 'Cells_AreaShape_Zernike_5_3', 'Cells_AreaShape_Zernike_5_5', 'Cells_AreaShape_Zernike_6_0', 'Cells_AreaShape_Zernike_6_2', 'Cells_AreaShape_Zernike_6_4', 'Cells_AreaShape_Zernike_6_6', 'Cells_AreaShape_Zernike_7_1', 'Cells_AreaShape_Zernike_7_3', 'Cells_AreaShape_Zernike_7_5', 'Cells_AreaShape_Zernike_7_7', 'Cells_AreaShape_Zernike_8_0', 'Cells_AreaShape_Zernike_8_2', 'Cells_AreaShape_Zernike_8_4', 'Cells_AreaShape_Zernike_8_6', 'Cells_AreaShape_Zernike_8_8', 'Cells_AreaShape_Zernike_9_1', 'Cells_AreaShape_Zernike_9_3', 'Cells_AreaShape_Zernike_9_5', 'Cells_AreaShape_Zernike_9_7', 'Cells_AreaShape_Zernike_9_9', 'Cells_Children_Cytoplasm_Count', 'Cells_Correlation_Correlation_DNA_AGP', 'Cells_Correlation_Correlation_DNA_ER', 'Cells_Correlation_Correlation_DNA_Mito', 'Cells_Correlation_Correlation_DNA_RNA', 'Cells_Correlation_Correlation_ER_AGP', 'Cells_Correlation_Correlation_ER_RNA', 'Cells_Correlation_Correlation_Mito_AGP', 'Cells_Correlation_Correlation_Mito_ER', 'Cells_Correlation_Correlation_Mito_RNA', 'Cells_Correlation_Correlation_RNA_AGP', 'Cells_Correlation_Costes_AGP_DNA', 'Cells_Correlation_Costes_AGP_ER', 'Cells_Correlation_Costes_AGP_Mito', 'Cells_Correlation_Costes_AGP_RNA', 'Cells_Correlation_Costes_DNA_AGP', 'Cells_Correlation_Costes_DNA_ER', 'Cells_Correlation_Costes_DNA_Mito', 'Cells_Correlation_Costes_DNA_RNA', 'Cells_Correlation_Costes_ER_AGP', 'Cells_Correlation_Costes_ER_DNA', 'Cells_Correlation_Costes_ER_Mito', 'Cells_Correlation_Costes_ER_RNA', 'Cells_Correlation_Costes_Mito_AGP', 'Cells_Correlation_Costes_Mito_DNA', 'Cells_Correlation_Costes_Mito_ER', 'Cells_Correlation_Costes_Mito_RNA', 'Cells_Correlation_Costes_RNA_AGP', 'Cells_Correlation_Costes_RNA_DNA', 'Cells_Correlation_Costes_RNA_ER', 'Cells_Correlation_Costes_RNA_Mito', 'Cells_Correlation_K_AGP_DNA', 'Cells_Correlation_K_AGP_ER', 'Cells_Correlation_K_AGP_Mito', 'Cells_Correlation_K_AGP_RNA', 'Cells_Correlation_K_DNA_AGP', 'Cells_Correlation_K_DNA_ER', 'Cells_Correlation_K_DNA_Mito', 'Cells_Correlation_K_DNA_RNA', 'Cells_Correlation_K_ER_AGP', 'Cells_Correlation_K_ER_DNA', 'Cells_Correlation_K_ER_Mito', 'Cells_Correlation_K_ER_RNA', 'Cells_Correlation_K_Mito_AGP', 'Cells_Correlation_K_Mito_DNA', 'Cells_Correlation_K_Mito_ER', 'Cells_Correlation_K_Mito_RNA', 'Cells_Correlation_K_RNA_AGP', 'Cells_Correlation_K_RNA_DNA', 'Cells_Correlation_K_RNA_ER', 'Cells_Correlation_K_RNA_Mito', 'Cells_Correlation_Manders_AGP_DNA', 'Cells_Correlation_Manders_AGP_ER', 'Cells_Correlation_Manders_AGP_Mito', 'Cells_Correlation_Manders_AGP_RNA', 'Cells_Correlation_Manders_DNA_AGP', 'Cells_Correlation_Manders_DNA_ER', 'Cells_Correlation_Manders_DNA_Mito', 'Cells_Correlation_Manders_DNA_RNA', 'Cells_Correlation_Manders_ER_AGP', 'Cells_Correlation_Manders_ER_DNA', 'Cells_Correlation_Manders_ER_Mito', 'Cells_Correlation_Manders_ER_RNA', 'Cells_Correlation_Manders_Mito_AGP', 'Cells_Correlation_Manders_Mito_DNA', 'Cells_Correlation_Manders_Mito_ER', 'Cells_Correlation_Manders_Mito_RNA', 'Cells_Correlation_Manders_RNA_AGP', 'Cells_Correlation_Manders_RNA_DNA', 'Cells_Correlation_Manders_RNA_ER', 'Cells_Correlation_Manders_RNA_Mito', 'Cells_Correlation_Overlap_DNA_AGP', 'Cells_Correlation_Overlap_DNA_ER', 'Cells_Correlation_Overlap_DNA_Mito', 'Cells_Correlation_Overlap_DNA_RNA', 'Cells_Correlation_Overlap_ER_AGP', 'Cells_Correlation_Overlap_ER_RNA', 'Cells_Correlation_Overlap_Mito_AGP', 'Cells_Correlation_Overlap_Mito_ER', 'Cells_Correlation_Overlap_Mito_RNA', 'Cells_Correlation_Overlap_RNA_AGP', 'Cells_Correlation_RWC_AGP_DNA', 'Cells_Correlation_RWC_AGP_ER', 'Cells_Correlation_RWC_AGP_Mito', 'Cells_Correlation_RWC_AGP_RNA', 'Cells_Correlation_RWC_DNA_AGP', 'Cells_Correlation_RWC_DNA_ER', 'Cells_Correlation_RWC_DNA_Mito', 'Cells_Correlation_RWC_DNA_RNA', 'Cells_Correlation_RWC_ER_AGP', 'Cells_Correlation_RWC_ER_DNA', 'Cells_Correlation_RWC_ER_Mito', 'Cells_Correlation_RWC_ER_RNA', 'Cells_Correlation_RWC_Mito_AGP', 'Cells_Correlation_RWC_Mito_DNA', 'Cells_Correlation_RWC_Mito_ER', 'Cells_Correlation_RWC_Mito_RNA', 'Cells_Correlation_RWC_RNA_AGP', 'Cells_Correlation_RWC_RNA_DNA', 'Cells_Correlation_RWC_RNA_ER', 'Cells_Correlation_RWC_RNA_Mito', 'Cells_Granularity_10_AGP', 'Cells_Granularity_10_ER', 'Cells_Granularity_10_Mito', 'Cells_Granularity_10_RNA', 'Cells_Granularity_11_AGP', 'Cells_Granularity_11_ER', 'Cells_Granularity_11_Mito', 'Cells_Granularity_11_RNA', 'Cells_Granularity_12_AGP', 'Cells_Granularity_12_ER', 'Cells_Granularity_12_Mito', 'Cells_Granularity_12_RNA', 'Cells_Granularity_13_AGP', 'Cells_Granularity_13_ER', 'Cells_Granularity_13_Mito', 'Cells_Granularity_13_RNA', 'Cells_Granularity_14_AGP', 'Cells_Granularity_14_ER', 'Cells_Granularity_14_Mito', 'Cells_Granularity_14_RNA', 'Cells_Granularity_15_AGP', 'Cells_Granularity_15_ER', 'Cells_Granularity_15_Mito', 'Cells_Granularity_15_RNA', 'Cells_Granularity_16_AGP', 'Cells_Granularity_16_ER', 'Cells_Granularity_16_Mito', 'Cells_Granularity_16_RNA', 'Cells_Granularity_1_AGP', 'Cells_Granularity_1_ER', 'Cells_Granularity_1_Mito', 'Cells_Granularity_1_RNA', 'Cells_Granularity_2_AGP', 'Cells_Granularity_2_ER', 'Cells_Granularity_2_Mito', 'Cells_Granularity_2_RNA', 'Cells_Granularity_3_AGP', 'Cells_Granularity_3_ER', 'Cells_Granularity_3_Mito', 'Cells_Granularity_3_RNA', 'Cells_Granularity_4_AGP', 'Cells_Granularity_4_ER', 'Cells_Granularity_4_Mito', 'Cells_Granularity_4_RNA', 'Cells_Granularity_5_AGP', 'Cells_Granularity_5_ER', 'Cells_Granularity_5_Mito', 'Cells_Granularity_5_RNA', 'Cells_Granularity_6_AGP', 'Cells_Granularity_6_ER', 'Cells_Granularity_6_Mito', 'Cells_Granularity_6_RNA', 'Cells_Granularity_7_AGP', 'Cells_Granularity_7_ER', 'Cells_Granularity_7_Mito', 'Cells_Granularity_7_RNA', 'Cells_Granularity_8_AGP', 'Cells_Granularity_8_ER', 'Cells_Granularity_8_Mito', 'Cells_Granularity_8_RNA', 'Cells_Granularity_9_AGP', 'Cells_Granularity_9_ER', 'Cells_Granularity_9_Mito', 'Cells_Granularity_9_RNA', 'Cells_Intensity_IntegratedIntensityEdge_AGP', 'Cells_Intensity_IntegratedIntensityEdge_DNA', 'Cells_Intensity_IntegratedIntensityEdge_ER', 'Cells_Intensity_IntegratedIntensityEdge_Mito', 'Cells_Intensity_IntegratedIntensityEdge_RNA', 'Cells_Intensity_IntegratedIntensity_AGP', 'Cells_Intensity_IntegratedIntensity_DNA', 'Cells_Intensity_IntegratedIntensity_ER', 'Cells_Intensity_IntegratedIntensity_Mito', 'Cells_Intensity_IntegratedIntensity_RNA', 'Cells_Intensity_LowerQuartileIntensity_AGP', 'Cells_Intensity_LowerQuartileIntensity_DNA', 'Cells_Intensity_LowerQuartileIntensity_ER', 'Cells_Intensity_LowerQuartileIntensity_Mito', 'Cells_Intensity_LowerQuartileIntensity_RNA', 'Cells_Intensity_MADIntensity_AGP', 'Cells_Intensity_MADIntensity_DNA', 'Cells_Intensity_MADIntensity_ER', 'Cells_Intensity_MADIntensity_Mito', 'Cells_Intensity_MADIntensity_RNA', 'Cells_Intensity_MassDisplacement_AGP', 'Cells_Intensity_MassDisplacement_DNA', 'Cells_Intensity_MassDisplacement_ER', 'Cells_Intensity_MassDisplacement_Mito', 'Cells_Intensity_MassDisplacement_RNA', 'Cells_Intensity_MaxIntensityEdge_AGP', 'Cells_Intensity_MaxIntensityEdge_DNA', 'Cells_Intensity_MaxIntensityEdge_ER', 'Cells_Intensity_MaxIntensityEdge_Mito', 'Cells_Intensity_MaxIntensityEdge_RNA', 'Cells_Intensity_MaxIntensity_AGP', 'Cells_Intensity_MaxIntensity_DNA', 'Cells_Intensity_MaxIntensity_ER', 'Cells_Intensity_MaxIntensity_Mito', 'Cells_Intensity_MaxIntensity_RNA', 'Cells_Intensity_MeanIntensityEdge_AGP', 'Cells_Intensity_MeanIntensityEdge_DNA', 'Cells_Intensity_MeanIntensityEdge_ER', 'Cells_Intensity_MeanIntensityEdge_Mito', 'Cells_Intensity_MeanIntensityEdge_RNA', 'Cells_Intensity_MeanIntensity_AGP', 'Cells_Intensity_MeanIntensity_DNA', 'Cells_Intensity_MeanIntensity_ER', 'Cells_Intensity_MeanIntensity_Mito', 'Cells_Intensity_MeanIntensity_RNA', 'Cells_Intensity_MedianIntensity_AGP', 'Cells_Intensity_MedianIntensity_DNA', 'Cells_Intensity_MedianIntensity_ER', 'Cells_Intensity_MedianIntensity_Mito', 'Cells_Intensity_MedianIntensity_RNA', 'Cells_Intensity_MinIntensityEdge_AGP', 'Cells_Intensity_MinIntensityEdge_DNA', 'Cells_Intensity_MinIntensityEdge_ER', 'Cells_Intensity_MinIntensityEdge_Mito', 'Cells_Intensity_MinIntensityEdge_RNA', 'Cells_Intensity_MinIntensity_AGP', 'Cells_Intensity_MinIntensity_DNA', 'Cells_Intensity_MinIntensity_ER', 'Cells_Intensity_MinIntensity_Mito', 'Cells_Intensity_MinIntensity_RNA', 'Cells_Intensity_StdIntensityEdge_AGP', 'Cells_Intensity_StdIntensityEdge_DNA', 'Cells_Intensity_StdIntensityEdge_ER', 'Cells_Intensity_StdIntensityEdge_Mito', 'Cells_Intensity_StdIntensityEdge_RNA', 'Cells_Intensity_StdIntensity_AGP', 'Cells_Intensity_StdIntensity_DNA', 'Cells_Intensity_StdIntensity_ER', 'Cells_Intensity_StdIntensity_Mito', 'Cells_Intensity_StdIntensity_RNA', 'Cells_Intensity_UpperQuartileIntensity_AGP', 'Cells_Intensity_UpperQuartileIntensity_DNA', 'Cells_Intensity_UpperQuartileIntensity_ER', 'Cells_Intensity_UpperQuartileIntensity_Mito', 'Cells_Intensity_UpperQuartileIntensity_RNA', 'Cells_Location_CenterMassIntensity_X_AGP', 'Cells_Location_CenterMassIntensity_X_DNA', 'Cells_Location_CenterMassIntensity_X_ER', 'Cells_Location_CenterMassIntensity_X_Mito', 'Cells_Location_CenterMassIntensity_X_RNA', 'Cells_Location_CenterMassIntensity_Y_AGP', 'Cells_Location_CenterMassIntensity_Y_DNA', 'Cells_Location_CenterMassIntensity_Y_ER', 'Cells_Location_CenterMassIntensity_Y_Mito', 'Cells_Location_CenterMassIntensity_Y_RNA', 'Cells_Location_Center_X', 'Cells_Location_Center_Y', 'Cells_Location_MaxIntensity_X_AGP', 'Cells_Location_MaxIntensity_X_DNA', 'Cells_Location_MaxIntensity_X_ER', 'Cells_Location_MaxIntensity_X_Mito', 'Cells_Location_MaxIntensity_X_RNA', 'Cells_Location_MaxIntensity_Y_AGP', 'Cells_Location_MaxIntensity_Y_DNA', 'Cells_Location_MaxIntensity_Y_ER', 'Cells_Location_MaxIntensity_Y_Mito', 'Cells_Location_MaxIntensity_Y_RNA', 'Cells_Neighbors_AngleBetweenNeighbors_5', 'Cells_Neighbors_AngleBetweenNeighbors_Adjacent', 'Cells_Neighbors_FirstClosestDistance_5', 'Cells_Neighbors_FirstClosestDistance_Adjacent', 'Cells_Neighbors_FirstClosestObjectNumber_5', 'Cells_Neighbors_FirstClosestObjectNumber_Adjacent', 'Cells_Neighbors_NumberOfNeighbors_5', 'Cells_Neighbors_NumberOfNeighbors_Adjacent', 'Cells_Neighbors_PercentTouching_5', 'Cells_Neighbors_PercentTouching_Adjacent', 'Cells_Neighbors_SecondClosestDistance_5', 'Cells_Neighbors_SecondClosestDistance_Adjacent', 'Cells_Neighbors_SecondClosestObjectNumber_5', 'Cells_Neighbors_SecondClosestObjectNumber_Adjacent', 'Cells_Number_Object_Number', 'Cells_Parent_Nuclei', 'Cells_RadialDistribution_FracAtD_AGP_1of4', 'Cells_RadialDistribution_FracAtD_AGP_2of4', 'Cells_RadialDistribution_FracAtD_AGP_3of4', 'Cells_RadialDistribution_FracAtD_AGP_4of4', 'Cells_RadialDistribution_FracAtD_DNA_1of4', 'Cells_RadialDistribution_FracAtD_DNA_2of4', 'Cells_RadialDistribution_FracAtD_DNA_3of4', 'Cells_RadialDistribution_FracAtD_DNA_4of4', 'Cells_RadialDistribution_FracAtD_ER_1of4', 'Cells_RadialDistribution_FracAtD_ER_2of4', 'Cells_RadialDistribution_FracAtD_ER_3of4', 'Cells_RadialDistribution_FracAtD_ER_4of4', 'Cells_RadialDistribution_FracAtD_Mito_1of4', 'Cells_RadialDistribution_FracAtD_Mito_2of4', 'Cells_RadialDistribution_FracAtD_Mito_3of4', 'Cells_RadialDistribution_FracAtD_Mito_4of4', 'Cells_RadialDistribution_FracAtD_RNA_1of4', 'Cells_RadialDistribution_FracAtD_RNA_2of4', 'Cells_RadialDistribution_FracAtD_RNA_3of4', 'Cells_RadialDistribution_FracAtD_RNA_4of4', 'Cells_RadialDistribution_MeanFrac_AGP_1of4', 'Cells_RadialDistribution_MeanFrac_AGP_2of4', 'Cells_RadialDistribution_MeanFrac_AGP_3of4', 'Cells_RadialDistribution_MeanFrac_AGP_4of4', 'Cells_RadialDistribution_MeanFrac_DNA_1of4', 'Cells_RadialDistribution_MeanFrac_DNA_2of4', 'Cells_RadialDistribution_MeanFrac_DNA_3of4', 'Cells_RadialDistribution_MeanFrac_DNA_4of4', 'Cells_RadialDistribution_MeanFrac_ER_1of4', 'Cells_RadialDistribution_MeanFrac_ER_2of4', 'Cells_RadialDistribution_MeanFrac_ER_3of4', 'Cells_RadialDistribution_MeanFrac_ER_4of4', 'Cells_RadialDistribution_MeanFrac_Mito_1of4', 'Cells_RadialDistribution_MeanFrac_Mito_2of4', 'Cells_RadialDistribution_MeanFrac_Mito_3of4', 'Cells_RadialDistribution_MeanFrac_Mito_4of4', 'Cells_RadialDistribution_MeanFrac_RNA_1of4', 'Cells_RadialDistribution_MeanFrac_RNA_2of4', 'Cells_RadialDistribution_MeanFrac_RNA_3of4', 'Cells_RadialDistribution_MeanFrac_RNA_4of4', 'Cells_RadialDistribution_RadialCV_AGP_1of4', 'Cells_RadialDistribution_RadialCV_AGP_2of4', 'Cells_RadialDistribution_RadialCV_AGP_3of4', 'Cells_RadialDistribution_RadialCV_AGP_4of4', 'Cells_RadialDistribution_RadialCV_DNA_1of4', 'Cells_RadialDistribution_RadialCV_DNA_2of4', 'Cells_RadialDistribution_RadialCV_DNA_3of4', 'Cells_RadialDistribution_RadialCV_DNA_4of4', 'Cells_RadialDistribution_RadialCV_ER_1of4', 'Cells_RadialDistribution_RadialCV_ER_2of4', 'Cells_RadialDistribution_RadialCV_ER_3of4', 'Cells_RadialDistribution_RadialCV_ER_4of4', 'Cells_RadialDistribution_RadialCV_Mito_1of4', 'Cells_RadialDistribution_RadialCV_Mito_2of4', 'Cells_RadialDistribution_RadialCV_Mito_3of4', 'Cells_RadialDistribution_RadialCV_Mito_4of4', 'Cells_RadialDistribution_RadialCV_RNA_1of4', 'Cells_RadialDistribution_RadialCV_RNA_2of4', 'Cells_RadialDistribution_RadialCV_RNA_3of4', 'Cells_RadialDistribution_RadialCV_RNA_4of4', 'Cells_Texture_AngularSecondMoment_AGP_10_0', 'Cells_Texture_AngularSecondMoment_AGP_3_0', 'Cells_Texture_AngularSecondMoment_AGP_5_0', 'Cells_Texture_AngularSecondMoment_DNA_10_0', 'Cells_Texture_AngularSecondMoment_DNA_3_0', 'Cells_Texture_AngularSecondMoment_DNA_5_0', 'Cells_Texture_AngularSecondMoment_ER_10_0', 'Cells_Texture_AngularSecondMoment_ER_3_0', 'Cells_Texture_AngularSecondMoment_ER_5_0', 'Cells_Texture_AngularSecondMoment_Mito_10_0', 'Cells_Texture_AngularSecondMoment_Mito_3_0', 'Cells_Texture_AngularSecondMoment_Mito_5_0', 'Cells_Texture_AngularSecondMoment_RNA_10_0', 'Cells_Texture_AngularSecondMoment_RNA_3_0', 'Cells_Texture_AngularSecondMoment_RNA_5_0', 'Cells_Texture_Contrast_AGP_10_0', 'Cells_Texture_Contrast_AGP_3_0', 'Cells_Texture_Contrast_AGP_5_0', 'Cells_Texture_Contrast_DNA_10_0', 'Cells_Texture_Contrast_DNA_3_0', 'Cells_Texture_Contrast_DNA_5_0', 'Cells_Texture_Contrast_ER_10_0', 'Cells_Texture_Contrast_ER_3_0', 'Cells_Texture_Contrast_ER_5_0', 'Cells_Texture_Contrast_Mito_10_0', 'Cells_Texture_Contrast_Mito_3_0', 'Cells_Texture_Contrast_Mito_5_0', 'Cells_Texture_Contrast_RNA_10_0', 'Cells_Texture_Contrast_RNA_3_0', 'Cells_Texture_Contrast_RNA_5_0', 'Cells_Texture_Correlation_AGP_10_0', 'Cells_Texture_Correlation_AGP_3_0', 'Cells_Texture_Correlation_AGP_5_0', 'Cells_Texture_Correlation_DNA_10_0', 'Cells_Texture_Correlation_DNA_3_0', 'Cells_Texture_Correlation_DNA_5_0', 'Cells_Texture_Correlation_ER_10_0', 'Cells_Texture_Correlation_ER_3_0', 'Cells_Texture_Correlation_ER_5_0', 'Cells_Texture_Correlation_Mito_10_0', 'Cells_Texture_Correlation_Mito_3_0', 'Cells_Texture_Correlation_Mito_5_0', 'Cells_Texture_Correlation_RNA_10_0', 'Cells_Texture_Correlation_RNA_3_0', 'Cells_Texture_Correlation_RNA_5_0', 'Cells_Texture_DifferenceEntropy_AGP_10_0', 'Cells_Texture_DifferenceEntropy_AGP_3_0', 'Cells_Texture_DifferenceEntropy_AGP_5_0', 'Cells_Texture_DifferenceEntropy_DNA_10_0', 'Cells_Texture_DifferenceEntropy_DNA_3_0', 'Cells_Texture_DifferenceEntropy_DNA_5_0', 'Cells_Texture_DifferenceEntropy_ER_10_0', 'Cells_Texture_DifferenceEntropy_ER_3_0', 'Cells_Texture_DifferenceEntropy_ER_5_0', 'Cells_Texture_DifferenceEntropy_Mito_10_0', 'Cells_Texture_DifferenceEntropy_Mito_3_0', 'Cells_Texture_DifferenceEntropy_Mito_5_0', 'Cells_Texture_DifferenceEntropy_RNA_10_0', 'Cells_Texture_DifferenceEntropy_RNA_3_0', 'Cells_Texture_DifferenceEntropy_RNA_5_0', 'Cells_Texture_DifferenceVariance_AGP_10_0', 'Cells_Texture_DifferenceVariance_AGP_3_0', 'Cells_Texture_DifferenceVariance_AGP_5_0', 'Cells_Texture_DifferenceVariance_DNA_10_0', 'Cells_Texture_DifferenceVariance_DNA_3_0', 'Cells_Texture_DifferenceVariance_DNA_5_0', 'Cells_Texture_DifferenceVariance_ER_10_0', 'Cells_Texture_DifferenceVariance_ER_3_0', 'Cells_Texture_DifferenceVariance_ER_5_0', 'Cells_Texture_DifferenceVariance_Mito_10_0', 'Cells_Texture_DifferenceVariance_Mito_3_0', 'Cells_Texture_DifferenceVariance_Mito_5_0', 'Cells_Texture_DifferenceVariance_RNA_10_0', 'Cells_Texture_DifferenceVariance_RNA_3_0', 'Cells_Texture_DifferenceVariance_RNA_5_0', 'Cells_Texture_Entropy_AGP_10_0', 'Cells_Texture_Entropy_AGP_3_0', 'Cells_Texture_Entropy_AGP_5_0', 'Cells_Texture_Entropy_DNA_10_0', 'Cells_Texture_Entropy_DNA_3_0', 'Cells_Texture_Entropy_DNA_5_0', 'Cells_Texture_Entropy_ER_10_0', 'Cells_Texture_Entropy_ER_3_0', 'Cells_Texture_Entropy_ER_5_0', 'Cells_Texture_Entropy_Mito_10_0', 'Cells_Texture_Entropy_Mito_3_0', 'Cells_Texture_Entropy_Mito_5_0', 'Cells_Texture_Entropy_RNA_10_0', 'Cells_Texture_Entropy_RNA_3_0', 'Cells_Texture_Entropy_RNA_5_0', 'Cells_Texture_Gabor_AGP_10', 'Cells_Texture_Gabor_AGP_3', 'Cells_Texture_Gabor_AGP_5', 'Cells_Texture_Gabor_DNA_10', 'Cells_Texture_Gabor_DNA_3', 'Cells_Texture_Gabor_DNA_5', 'Cells_Texture_Gabor_ER_10', 'Cells_Texture_Gabor_ER_3', 'Cells_Texture_Gabor_ER_5', 'Cells_Texture_Gabor_Mito_10', 'Cells_Texture_Gabor_Mito_3', 'Cells_Texture_Gabor_Mito_5', 'Cells_Texture_Gabor_RNA_10', 'Cells_Texture_Gabor_RNA_3', 'Cells_Texture_Gabor_RNA_5', 'Cells_Texture_InfoMeas1_AGP_10_0', 'Cells_Texture_InfoMeas1_AGP_3_0', 'Cells_Texture_InfoMeas1_AGP_5_0', 'Cells_Texture_InfoMeas1_DNA_10_0', 'Cells_Texture_InfoMeas1_DNA_3_0', 'Cells_Texture_InfoMeas1_DNA_5_0', 'Cells_Texture_InfoMeas1_ER_10_0', 'Cells_Texture_InfoMeas1_ER_3_0', 'Cells_Texture_InfoMeas1_ER_5_0', 'Cells_Texture_InfoMeas1_Mito_10_0', 'Cells_Texture_InfoMeas1_Mito_3_0', 'Cells_Texture_InfoMeas1_Mito_5_0', 'Cells_Texture_InfoMeas1_RNA_10_0', 'Cells_Texture_InfoMeas1_RNA_3_0', 'Cells_Texture_InfoMeas1_RNA_5_0', 'Cells_Texture_InfoMeas2_AGP_10_0', 'Cells_Texture_InfoMeas2_AGP_3_0', 'Cells_Texture_InfoMeas2_AGP_5_0', 'Cells_Texture_InfoMeas2_DNA_10_0', 'Cells_Texture_InfoMeas2_DNA_3_0', 'Cells_Texture_InfoMeas2_DNA_5_0', 'Cells_Texture_InfoMeas2_ER_10_0', 'Cells_Texture_InfoMeas2_ER_3_0', 'Cells_Texture_InfoMeas2_ER_5_0', 'Cells_Texture_InfoMeas2_Mito_10_0', 'Cells_Texture_InfoMeas2_Mito_3_0', 'Cells_Texture_InfoMeas2_Mito_5_0', 'Cells_Texture_InfoMeas2_RNA_10_0', 'Cells_Texture_InfoMeas2_RNA_3_0', 'Cells_Texture_InfoMeas2_RNA_5_0', 'Cells_Texture_InverseDifferenceMoment_AGP_10_0', 'Cells_Texture_InverseDifferenceMoment_AGP_3_0', 'Cells_Texture_InverseDifferenceMoment_AGP_5_0', 'Cells_Texture_InverseDifferenceMoment_DNA_10_0', 'Cells_Texture_InverseDifferenceMoment_DNA_3_0', 'Cells_Texture_InverseDifferenceMoment_DNA_5_0', 'Cells_Texture_InverseDifferenceMoment_ER_10_0', 'Cells_Texture_InverseDifferenceMoment_ER_3_0', 'Cells_Texture_InverseDifferenceMoment_ER_5_0', 'Cells_Texture_InverseDifferenceMoment_Mito_10_0', 'Cells_Texture_InverseDifferenceMoment_Mito_3_0', 'Cells_Texture_InverseDifferenceMoment_Mito_5_0', 'Cells_Texture_InverseDifferenceMoment_RNA_10_0', 'Cells_Texture_InverseDifferenceMoment_RNA_3_0', 'Cells_Texture_InverseDifferenceMoment_RNA_5_0', 'Cells_Texture_SumAverage_AGP_10_0', 'Cells_Texture_SumAverage_AGP_3_0', 'Cells_Texture_SumAverage_AGP_5_0', 'Cells_Texture_SumAverage_DNA_10_0', 'Cells_Texture_SumAverage_DNA_3_0', 'Cells_Texture_SumAverage_DNA_5_0', 'Cells_Texture_SumAverage_ER_10_0', 'Cells_Texture_SumAverage_ER_3_0', 'Cells_Texture_SumAverage_ER_5_0', 'Cells_Texture_SumAverage_Mito_10_0', 'Cells_Texture_SumAverage_Mito_3_0', 'Cells_Texture_SumAverage_Mito_5_0', 'Cells_Texture_SumAverage_RNA_10_0', 'Cells_Texture_SumAverage_RNA_3_0', 'Cells_Texture_SumAverage_RNA_5_0', 'Cells_Texture_SumEntropy_AGP_10_0', 'Cells_Texture_SumEntropy_AGP_3_0', 'Cells_Texture_SumEntropy_AGP_5_0', 'Cells_Texture_SumEntropy_DNA_10_0', 'Cells_Texture_SumEntropy_DNA_3_0', 'Cells_Texture_SumEntropy_DNA_5_0', 'Cells_Texture_SumEntropy_ER_10_0', 'Cells_Texture_SumEntropy_ER_3_0', 'Cells_Texture_SumEntropy_ER_5_0', 'Cells_Texture_SumEntropy_Mito_10_0', 'Cells_Texture_SumEntropy_Mito_3_0', 'Cells_Texture_SumEntropy_Mito_5_0', 'Cells_Texture_SumEntropy_RNA_10_0', 'Cells_Texture_SumEntropy_RNA_3_0', 'Cells_Texture_SumEntropy_RNA_5_0', 'Cells_Texture_SumVariance_AGP_10_0', 'Cells_Texture_SumVariance_AGP_3_0', 'Cells_Texture_SumVariance_AGP_5_0', 'Cells_Texture_SumVariance_DNA_10_0', 'Cells_Texture_SumVariance_DNA_3_0', 'Cells_Texture_SumVariance_DNA_5_0', 'Cells_Texture_SumVariance_ER_10_0', 'Cells_Texture_SumVariance_ER_3_0', 'Cells_Texture_SumVariance_ER_5_0', 'Cells_Texture_SumVariance_Mito_10_0', 'Cells_Texture_SumVariance_Mito_3_0', 'Cells_Texture_SumVariance_Mito_5_0', 'Cells_Texture_SumVariance_RNA_10_0', 'Cells_Texture_SumVariance_RNA_3_0', 'Cells_Texture_SumVariance_RNA_5_0', 'Cells_Texture_Variance_AGP_10_0', 'Cells_Texture_Variance_AGP_3_0', 'Cells_Texture_Variance_AGP_5_0', 'Cells_Texture_Variance_DNA_10_0', 'Cells_Texture_Variance_DNA_3_0', 'Cells_Texture_Variance_DNA_5_0', 'Cells_Texture_Variance_ER_10_0', 'Cells_Texture_Variance_ER_3_0', 'Cells_Texture_Variance_ER_5_0', 'Cells_Texture_Variance_Mito_10_0', 'Cells_Texture_Variance_Mito_3_0', 'Cells_Texture_Variance_Mito_5_0', 'Cells_Texture_Variance_RNA_10_0', 'Cells_Texture_Variance_RNA_3_0', 'Cells_Texture_Variance_RNA_5_0'

Cytoplasm Table

'Cytoplasm_AreaShape_Area', 'Cytoplasm_AreaShape_Center_X', 'Cytoplasm_AreaShape_Center_Y', 'Cytoplasm_AreaShape_Compactness', 'Cytoplasm_AreaShape_Eccentricity', 'Cytoplasm_AreaShape_EulerNumber', 'Cytoplasm_AreaShape_Extent', 'Cytoplasm_AreaShape_FormFactor', 'Cytoplasm_AreaShape_MajorAxisLength', 'Cytoplasm_AreaShape_MaxFeretDiameter', 'Cytoplasm_AreaShape_MaximumRadius', 'Cytoplasm_AreaShape_MeanRadius', 'Cytoplasm_AreaShape_MedianRadius', 'Cytoplasm_AreaShape_MinFeretDiameter', 'Cytoplasm_AreaShape_MinorAxisLength', 'Cytoplasm_AreaShape_Orientation', 'Cytoplasm_AreaShape_Perimeter', 'Cytoplasm_AreaShape_Solidity', 'Cytoplasm_AreaShape_Zernike_0_0', 'Cytoplasm_AreaShape_Zernike_1_1', 'Cytoplasm_AreaShape_Zernike_2_0', 'Cytoplasm_AreaShape_Zernike_2_2', 'Cytoplasm_AreaShape_Zernike_3_1', 'Cytoplasm_AreaShape_Zernike_3_3', 'Cytoplasm_AreaShape_Zernike_4_0', 'Cytoplasm_AreaShape_Zernike_4_2', 'Cytoplasm_AreaShape_Zernike_4_4', 'Cytoplasm_AreaShape_Zernike_5_1', 'Cytoplasm_AreaShape_Zernike_5_3', 'Cytoplasm_AreaShape_Zernike_5_5', 'Cytoplasm_AreaShape_Zernike_6_0', 'Cytoplasm_AreaShape_Zernike_6_2', 'Cytoplasm_AreaShape_Zernike_6_4', 'Cytoplasm_AreaShape_Zernike_6_6', 'Cytoplasm_AreaShape_Zernike_7_1', 'Cytoplasm_AreaShape_Zernike_7_3', 'Cytoplasm_AreaShape_Zernike_7_5', 'Cytoplasm_AreaShape_Zernike_7_7', 'Cytoplasm_AreaShape_Zernike_8_0', 'Cytoplasm_AreaShape_Zernike_8_2', 'Cytoplasm_AreaShape_Zernike_8_4', 'Cytoplasm_AreaShape_Zernike_8_6', 'Cytoplasm_AreaShape_Zernike_8_8', 'Cytoplasm_AreaShape_Zernike_9_1', 'Cytoplasm_AreaShape_Zernike_9_3', 'Cytoplasm_AreaShape_Zernike_9_5', 'Cytoplasm_AreaShape_Zernike_9_7', 'Cytoplasm_AreaShape_Zernike_9_9', 'Cytoplasm_Correlation_Correlation_DNA_AGP', 'Cytoplasm_Correlation_Correlation_DNA_ER', 'Cytoplasm_Correlation_Correlation_DNA_Mito', 'Cytoplasm_Correlation_Correlation_DNA_RNA', 'Cytoplasm_Correlation_Correlation_ER_AGP', 'Cytoplasm_Correlation_Correlation_ER_RNA', 'Cytoplasm_Correlation_Correlation_Mito_AGP', 'Cytoplasm_Correlation_Correlation_Mito_ER', 'Cytoplasm_Correlation_Correlation_Mito_RNA', 'Cytoplasm_Correlation_Correlation_RNA_AGP', 'Cytoplasm_Correlation_Costes_AGP_DNA', 'Cytoplasm_Correlation_Costes_AGP_ER', 'Cytoplasm_Correlation_Costes_AGP_Mito', 'Cytoplasm_Correlation_Costes_AGP_RNA', 'Cytoplasm_Correlation_Costes_DNA_AGP', 'Cytoplasm_Correlation_Costes_DNA_ER', 'Cytoplasm_Correlation_Costes_DNA_Mito', 'Cytoplasm_Correlation_Costes_DNA_RNA', 'Cytoplasm_Correlation_Costes_ER_AGP', 'Cytoplasm_Correlation_Costes_ER_DNA', 'Cytoplasm_Correlation_Costes_ER_Mito', 'Cytoplasm_Correlation_Costes_ER_RNA', 'Cytoplasm_Correlation_Costes_Mito_AGP', 'Cytoplasm_Correlation_Costes_Mito_DNA', 'Cytoplasm_Correlation_Costes_Mito_ER', 'Cytoplasm_Correlation_Costes_Mito_RNA', 'Cytoplasm_Correlation_Costes_RNA_AGP', 'Cytoplasm_Correlation_Costes_RNA_DNA', 'Cytoplasm_Correlation_Costes_RNA_ER', 'Cytoplasm_Correlation_Costes_RNA_Mito', 'Cytoplasm_Correlation_K_AGP_DNA', 'Cytoplasm_Correlation_K_AGP_ER', 'Cytoplasm_Correlation_K_AGP_Mito', 'Cytoplasm_Correlation_K_AGP_RNA', 'Cytoplasm_Correlation_K_DNA_AGP', 'Cytoplasm_Correlation_K_DNA_ER', 'Cytoplasm_Correlation_K_DNA_Mito', 'Cytoplasm_Correlation_K_DNA_RNA', 'Cytoplasm_Correlation_K_ER_AGP', 'Cytoplasm_Correlation_K_ER_DNA', 'Cytoplasm_Correlation_K_ER_Mito', 'Cytoplasm_Correlation_K_ER_RNA', 'Cytoplasm_Correlation_K_Mito_AGP', 'Cytoplasm_Correlation_K_Mito_DNA', 'Cytoplasm_Correlation_K_Mito_ER', 'Cytoplasm_Correlation_K_Mito_RNA', 'Cytoplasm_Correlation_K_RNA_AGP', 'Cytoplasm_Correlation_K_RNA_DNA', 'Cytoplasm_Correlation_K_RNA_ER', 'Cytoplasm_Correlation_K_RNA_Mito', 'Cytoplasm_Correlation_Manders_AGP_DNA', 'Cytoplasm_Correlation_Manders_AGP_ER', 'Cytoplasm_Correlation_Manders_AGP_Mito', 'Cytoplasm_Correlation_Manders_AGP_RNA', 'Cytoplasm_Correlation_Manders_DNA_AGP', 'Cytoplasm_Correlation_Manders_DNA_ER', 'Cytoplasm_Correlation_Manders_DNA_Mito', 'Cytoplasm_Correlation_Manders_DNA_RNA', 'Cytoplasm_Correlation_Manders_ER_AGP', 'Cytoplasm_Correlation_Manders_ER_DNA', 'Cytoplasm_Correlation_Manders_ER_Mito', 'Cytoplasm_Correlation_Manders_ER_RNA', 'Cytoplasm_Correlation_Manders_Mito_AGP', 'Cytoplasm_Correlation_Manders_Mito_DNA', 'Cytoplasm_Correlation_Manders_Mito_ER', 'Cytoplasm_Correlation_Manders_Mito_RNA', 'Cytoplasm_Correlation_Manders_RNA_AGP', 'Cytoplasm_Correlation_Manders_RNA_DNA', 'Cytoplasm_Correlation_Manders_RNA_ER', 'Cytoplasm_Correlation_Manders_RNA_Mito', 'Cytoplasm_Correlation_Overlap_DNA_AGP', 'Cytoplasm_Correlation_Overlap_DNA_ER', 'Cytoplasm_Correlation_Overlap_DNA_Mito', 'Cytoplasm_Correlation_Overlap_DNA_RNA', 'Cytoplasm_Correlation_Overlap_ER_AGP', 'Cytoplasm_Correlation_Overlap_ER_RNA', 'Cytoplasm_Correlation_Overlap_Mito_AGP', 'Cytoplasm_Correlation_Overlap_Mito_ER', 'Cytoplasm_Correlation_Overlap_Mito_RNA', 'Cytoplasm_Correlation_Overlap_RNA_AGP', 'Cytoplasm_Correlation_RWC_AGP_DNA', 'Cytoplasm_Correlation_RWC_AGP_ER', 'Cytoplasm_Correlation_RWC_AGP_Mito', 'Cytoplasm_Correlation_RWC_AGP_RNA', 'Cytoplasm_Correlation_RWC_DNA_AGP', 'Cytoplasm_Correlation_RWC_DNA_ER', 'Cytoplasm_Correlation_RWC_DNA_Mito', 'Cytoplasm_Correlation_RWC_DNA_RNA', 'Cytoplasm_Correlation_RWC_ER_AGP', 'Cytoplasm_Correlation_RWC_ER_DNA', 'Cytoplasm_Correlation_RWC_ER_Mito', 'Cytoplasm_Correlation_RWC_ER_RNA', 'Cytoplasm_Correlation_RWC_Mito_AGP', 'Cytoplasm_Correlation_RWC_Mito_DNA', 'Cytoplasm_Correlation_RWC_Mito_ER', 'Cytoplasm_Correlation_RWC_Mito_RNA', 'Cytoplasm_Correlation_RWC_RNA_AGP', 'Cytoplasm_Correlation_RWC_RNA_DNA', 'Cytoplasm_Correlation_RWC_RNA_ER', 'Cytoplasm_Correlation_RWC_RNA_Mito', 'Cytoplasm_Granularity_10_AGP', 'Cytoplasm_Granularity_10_ER', 'Cytoplasm_Granularity_10_Mito', 'Cytoplasm_Granularity_10_RNA', 'Cytoplasm_Granularity_11_AGP', 'Cytoplasm_Granularity_11_ER', 'Cytoplasm_Granularity_11_Mito', 'Cytoplasm_Granularity_11_RNA', 'Cytoplasm_Granularity_12_AGP', 'Cytoplasm_Granularity_12_ER', 'Cytoplasm_Granularity_12_Mito', 'Cytoplasm_Granularity_12_RNA', 'Cytoplasm_Granularity_13_AGP', 'Cytoplasm_Granularity_13_ER', 'Cytoplasm_Granularity_13_Mito', 'Cytoplasm_Granularity_13_RNA', 'Cytoplasm_Granularity_14_AGP', 'Cytoplasm_Granularity_14_ER', 'Cytoplasm_Granularity_14_Mito', 'Cytoplasm_Granularity_14_RNA', 'Cytoplasm_Granularity_15_AGP', 'Cytoplasm_Granularity_15_ER', 'Cytoplasm_Granularity_15_Mito', 'Cytoplasm_Granularity_15_RNA', 'Cytoplasm_Granularity_16_AGP', 'Cytoplasm_Granularity_16_ER', 'Cytoplasm_Granularity_16_Mito', 'Cytoplasm_Granularity_16_RNA', 'Cytoplasm_Granularity_1_AGP', 'Cytoplasm_Granularity_1_ER', 'Cytoplasm_Granularity_1_Mito', 'Cytoplasm_Granularity_1_RNA', 'Cytoplasm_Granularity_2_AGP', 'Cytoplasm_Granularity_2_ER', 'Cytoplasm_Granularity_2_Mito', 'Cytoplasm_Granularity_2_RNA', 'Cytoplasm_Granularity_3_AGP', 'Cytoplasm_Granularity_3_ER', 'Cytoplasm_Granularity_3_Mito', 'Cytoplasm_Granularity_3_RNA', 'Cytoplasm_Granularity_4_AGP', 'Cytoplasm_Granularity_4_ER', 'Cytoplasm_Granularity_4_Mito', 'Cytoplasm_Granularity_4_RNA', 'Cytoplasm_Granularity_5_AGP', 'Cytoplasm_Granularity_5_ER', 'Cytoplasm_Granularity_5_Mito', 'Cytoplasm_Granularity_5_RNA', 'Cytoplasm_Granularity_6_AGP', 'Cytoplasm_Granularity_6_ER', 'Cytoplasm_Granularity_6_Mito', 'Cytoplasm_Granularity_6_RNA', 'Cytoplasm_Granularity_7_AGP', 'Cytoplasm_Granularity_7_ER', 'Cytoplasm_Granularity_7_Mito', 'Cytoplasm_Granularity_7_RNA', 'Cytoplasm_Granularity_8_AGP', 'Cytoplasm_Granularity_8_ER', 'Cytoplasm_Granularity_8_Mito', 'Cytoplasm_Granularity_8_RNA', 'Cytoplasm_Granularity_9_AGP', 'Cytoplasm_Granularity_9_ER', 'Cytoplasm_Granularity_9_Mito', 'Cytoplasm_Granularity_9_RNA', 'Cytoplasm_Intensity_IntegratedIntensityEdge_AGP', 'Cytoplasm_Intensity_IntegratedIntensityEdge_DNA', 'Cytoplasm_Intensity_IntegratedIntensityEdge_ER', 'Cytoplasm_Intensity_IntegratedIntensityEdge_Mito', 'Cytoplasm_Intensity_IntegratedIntensityEdge_RNA', 'Cytoplasm_Intensity_IntegratedIntensity_AGP', 'Cytoplasm_Intensity_IntegratedIntensity_DNA', 'Cytoplasm_Intensity_IntegratedIntensity_ER', 'Cytoplasm_Intensity_IntegratedIntensity_Mito', 'Cytoplasm_Intensity_IntegratedIntensity_RNA', 'Cytoplasm_Intensity_LowerQuartileIntensity_AGP', 'Cytoplasm_Intensity_LowerQuartileIntensity_DNA', 'Cytoplasm_Intensity_LowerQuartileIntensity_ER', 'Cytoplasm_Intensity_LowerQuartileIntensity_Mito', 'Cytoplasm_Intensity_LowerQuartileIntensity_RNA', 'Cytoplasm_Intensity_MADIntensity_AGP', 'Cytoplasm_Intensity_MADIntensity_DNA', 'Cytoplasm_Intensity_MADIntensity_ER', 'Cytoplasm_Intensity_MADIntensity_Mito', 'Cytoplasm_Intensity_MADIntensity_RNA', 'Cytoplasm_Intensity_MassDisplacement_AGP', 'Cytoplasm_Intensity_MassDisplacement_DNA', 'Cytoplasm_Intensity_MassDisplacement_ER', 'Cytoplasm_Intensity_MassDisplacement_Mito', 'Cytoplasm_Intensity_MassDisplacement_RNA', 'Cytoplasm_Intensity_MaxIntensityEdge_AGP', 'Cytoplasm_Intensity_MaxIntensityEdge_DNA', 'Cytoplasm_Intensity_MaxIntensityEdge_ER', 'Cytoplasm_Intensity_MaxIntensityEdge_Mito', 'Cytoplasm_Intensity_MaxIntensityEdge_RNA', 'Cytoplasm_Intensity_MaxIntensity_AGP', 'Cytoplasm_Intensity_MaxIntensity_DNA', 'Cytoplasm_Intensity_MaxIntensity_ER', 'Cytoplasm_Intensity_MaxIntensity_Mito', 'Cytoplasm_Intensity_MaxIntensity_RNA', 'Cytoplasm_Intensity_MeanIntensityEdge_AGP', 'Cytoplasm_Intensity_MeanIntensityEdge_DNA', 'Cytoplasm_Intensity_MeanIntensityEdge_ER', 'Cytoplasm_Intensity_MeanIntensityEdge_Mito', 'Cytoplasm_Intensity_MeanIntensityEdge_RNA', 'Cytoplasm_Intensity_MeanIntensity_AGP', 'Cytoplasm_Intensity_MeanIntensity_DNA', 'Cytoplasm_Intensity_MeanIntensity_ER', 'Cytoplasm_Intensity_MeanIntensity_Mito', 'Cytoplasm_Intensity_MeanIntensity_RNA', 'Cytoplasm_Intensity_MedianIntensity_AGP', 'Cytoplasm_Intensity_MedianIntensity_DNA', 'Cytoplasm_Intensity_MedianIntensity_ER', 'Cytoplasm_Intensity_MedianIntensity_Mito', 'Cytoplasm_Intensity_MedianIntensity_RNA', 'Cytoplasm_Intensity_MinIntensityEdge_AGP', 'Cytoplasm_Intensity_MinIntensityEdge_DNA', 'Cytoplasm_Intensity_MinIntensityEdge_ER', 'Cytoplasm_Intensity_MinIntensityEdge_Mito', 'Cytoplasm_Intensity_MinIntensityEdge_RNA', 'Cytoplasm_Intensity_MinIntensity_AGP', 'Cytoplasm_Intensity_MinIntensity_DNA', 'Cytoplasm_Intensity_MinIntensity_ER', 'Cytoplasm_Intensity_MinIntensity_Mito', 'Cytoplasm_Intensity_MinIntensity_RNA', 'Cytoplasm_Intensity_StdIntensityEdge_AGP', 'Cytoplasm_Intensity_StdIntensityEdge_DNA', 'Cytoplasm_Intensity_StdIntensityEdge_ER', 'Cytoplasm_Intensity_StdIntensityEdge_Mito', 'Cytoplasm_Intensity_StdIntensityEdge_RNA', 'Cytoplasm_Intensity_StdIntensity_AGP', 'Cytoplasm_Intensity_StdIntensity_DNA', 'Cytoplasm_Intensity_StdIntensity_ER', 'Cytoplasm_Intensity_StdIntensity_Mito', 'Cytoplasm_Intensity_StdIntensity_RNA', 'Cytoplasm_Intensity_UpperQuartileIntensity_AGP', 'Cytoplasm_Intensity_UpperQuartileIntensity_DNA', 'Cytoplasm_Intensity_UpperQuartileIntensity_ER', 'Cytoplasm_Intensity_UpperQuartileIntensity_Mito', 'Cytoplasm_Intensity_UpperQuartileIntensity_RNA', 'Cytoplasm_Location_CenterMassIntensity_X_AGP', 'Cytoplasm_Location_CenterMassIntensity_X_DNA', 'Cytoplasm_Location_CenterMassIntensity_X_ER', 'Cytoplasm_Location_CenterMassIntensity_X_Mito', 'Cytoplasm_Location_CenterMassIntensity_X_RNA', 'Cytoplasm_Location_CenterMassIntensity_Y_AGP', 'Cytoplasm_Location_CenterMassIntensity_Y_DNA', 'Cytoplasm_Location_CenterMassIntensity_Y_ER', 'Cytoplasm_Location_CenterMassIntensity_Y_Mito', 'Cytoplasm_Location_CenterMassIntensity_Y_RNA', 'Cytoplasm_Location_Center_X', 'Cytoplasm_Location_Center_Y', 'Cytoplasm_Location_MaxIntensity_X_AGP', 'Cytoplasm_Location_MaxIntensity_X_DNA', 'Cytoplasm_Location_MaxIntensity_X_ER', 'Cytoplasm_Location_MaxIntensity_X_Mito', 'Cytoplasm_Location_MaxIntensity_X_RNA', 'Cytoplasm_Location_MaxIntensity_Y_AGP', 'Cytoplasm_Location_MaxIntensity_Y_DNA', 'Cytoplasm_Location_MaxIntensity_Y_ER', 'Cytoplasm_Location_MaxIntensity_Y_Mito', 'Cytoplasm_Location_MaxIntensity_Y_RNA', 'Cytoplasm_Number_Object_Number', 'Cytoplasm_Parent_Cells', 'Cytoplasm_Parent_Nuclei', 'Cytoplasm_RadialDistribution_FracAtD_AGP_1of4', 'Cytoplasm_RadialDistribution_FracAtD_AGP_2of4', 'Cytoplasm_RadialDistribution_FracAtD_AGP_3of4', 'Cytoplasm_RadialDistribution_FracAtD_AGP_4of4', 'Cytoplasm_RadialDistribution_FracAtD_DNA_1of4', 'Cytoplasm_RadialDistribution_FracAtD_DNA_2of4', 'Cytoplasm_RadialDistribution_FracAtD_DNA_3of4', 'Cytoplasm_RadialDistribution_FracAtD_DNA_4of4', 'Cytoplasm_RadialDistribution_FracAtD_ER_1of4', 'Cytoplasm_RadialDistribution_FracAtD_ER_2of4', 'Cytoplasm_RadialDistribution_FracAtD_ER_3of4', 'Cytoplasm_RadialDistribution_FracAtD_ER_4of4', 'Cytoplasm_RadialDistribution_FracAtD_Mito_1of4', 'Cytoplasm_RadialDistribution_FracAtD_Mito_2of4', 'Cytoplasm_RadialDistribution_FracAtD_Mito_3of4', 'Cytoplasm_RadialDistribution_FracAtD_Mito_4of4', 'Cytoplasm_RadialDistribution_FracAtD_RNA_1of4', 'Cytoplasm_RadialDistribution_FracAtD_RNA_2of4', 'Cytoplasm_RadialDistribution_FracAtD_RNA_3of4', 'Cytoplasm_RadialDistribution_FracAtD_RNA_4of4', 'Cytoplasm_RadialDistribution_MeanFrac_AGP_1of4', 'Cytoplasm_RadialDistribution_MeanFrac_AGP_2of4', 'Cytoplasm_RadialDistribution_MeanFrac_AGP_3of4', 'Cytoplasm_RadialDistribution_MeanFrac_AGP_4of4', 'Cytoplasm_RadialDistribution_MeanFrac_DNA_1of4', 'Cytoplasm_RadialDistribution_MeanFrac_DNA_2of4', 'Cytoplasm_RadialDistribution_MeanFrac_DNA_3of4', 'Cytoplasm_RadialDistribution_MeanFrac_DNA_4of4', 'Cytoplasm_RadialDistribution_MeanFrac_ER_1of4', 'Cytoplasm_RadialDistribution_MeanFrac_ER_2of4', 'Cytoplasm_RadialDistribution_MeanFrac_ER_3of4', 'Cytoplasm_RadialDistribution_MeanFrac_ER_4of4', 'Cytoplasm_RadialDistribution_MeanFrac_Mito_1of4', 'Cytoplasm_RadialDistribution_MeanFrac_Mito_2of4', 'Cytoplasm_RadialDistribution_MeanFrac_Mito_3of4', 'Cytoplasm_RadialDistribution_MeanFrac_Mito_4of4', 'Cytoplasm_RadialDistribution_MeanFrac_RNA_1of4', 'Cytoplasm_RadialDistribution_MeanFrac_RNA_2of4', 'Cytoplasm_RadialDistribution_MeanFrac_RNA_3of4', 'Cytoplasm_RadialDistribution_MeanFrac_RNA_4of4', 'Cytoplasm_RadialDistribution_RadialCV_AGP_1of4', 'Cytoplasm_RadialDistribution_RadialCV_AGP_2of4', 'Cytoplasm_RadialDistribution_RadialCV_AGP_3of4', 'Cytoplasm_RadialDistribution_RadialCV_AGP_4of4', 'Cytoplasm_RadialDistribution_RadialCV_DNA_1of4', 'Cytoplasm_RadialDistribution_RadialCV_DNA_2of4', 'Cytoplasm_RadialDistribution_RadialCV_DNA_3of4', 'Cytoplasm_RadialDistribution_RadialCV_DNA_4of4', 'Cytoplasm_RadialDistribution_RadialCV_ER_1of4', 'Cytoplasm_RadialDistribution_RadialCV_ER_2of4', 'Cytoplasm_RadialDistribution_RadialCV_ER_3of4', 'Cytoplasm_RadialDistribution_RadialCV_ER_4of4', 'Cytoplasm_RadialDistribution_RadialCV_Mito_1of4', 'Cytoplasm_RadialDistribution_RadialCV_Mito_2of4', 'Cytoplasm_RadialDistribution_RadialCV_Mito_3of4', 'Cytoplasm_RadialDistribution_RadialCV_Mito_4of4', 'Cytoplasm_RadialDistribution_RadialCV_RNA_1of4', 'Cytoplasm_RadialDistribution_RadialCV_RNA_2of4', 'Cytoplasm_RadialDistribution_RadialCV_RNA_3of4', 'Cytoplasm_RadialDistribution_RadialCV_RNA_4of4', 'Cytoplasm_Texture_AngularSecondMoment_AGP_10_0', 'Cytoplasm_Texture_AngularSecondMoment_AGP_3_0', 'Cytoplasm_Texture_AngularSecondMoment_AGP_5_0', 'Cytoplasm_Texture_AngularSecondMoment_DNA_10_0', 'Cytoplasm_Texture_AngularSecondMoment_DNA_3_0', 'Cytoplasm_Texture_AngularSecondMoment_DNA_5_0', 'Cytoplasm_Texture_AngularSecondMoment_ER_10_0', 'Cytoplasm_Texture_AngularSecondMoment_ER_3_0', 'Cytoplasm_Texture_AngularSecondMoment_ER_5_0', 'Cytoplasm_Texture_AngularSecondMoment_Mito_10_0', 'Cytoplasm_Texture_AngularSecondMoment_Mito_3_0', 'Cytoplasm_Texture_AngularSecondMoment_Mito_5_0', 'Cytoplasm_Texture_AngularSecondMoment_RNA_10_0', 'Cytoplasm_Texture_AngularSecondMoment_RNA_3_0', 'Cytoplasm_Texture_AngularSecondMoment_RNA_5_0', 'Cytoplasm_Texture_Contrast_AGP_10_0', 'Cytoplasm_Texture_Contrast_AGP_3_0', 'Cytoplasm_Texture_Contrast_AGP_5_0', 'Cytoplasm_Texture_Contrast_DNA_10_0', 'Cytoplasm_Texture_Contrast_DNA_3_0', 'Cytoplasm_Texture_Contrast_DNA_5_0', 'Cytoplasm_Texture_Contrast_ER_10_0', 'Cytoplasm_Texture_Contrast_ER_3_0', 'Cytoplasm_Texture_Contrast_ER_5_0', 'Cytoplasm_Texture_Contrast_Mito_10_0', 'Cytoplasm_Texture_Contrast_Mito_3_0', 'Cytoplasm_Texture_Contrast_Mito_5_0', 'Cytoplasm_Texture_Contrast_RNA_10_0', 'Cytoplasm_Texture_Contrast_RNA_3_0', 'Cytoplasm_Texture_Contrast_RNA_5_0', 'Cytoplasm_Texture_Correlation_AGP_10_0', 'Cytoplasm_Texture_Correlation_AGP_3_0', 'Cytoplasm_Texture_Correlation_AGP_5_0', 'Cytoplasm_Texture_Correlation_DNA_10_0', 'Cytoplasm_Texture_Correlation_DNA_3_0', 'Cytoplasm_Texture_Correlation_DNA_5_0', 'Cytoplasm_Texture_Correlation_ER_10_0', 'Cytoplasm_Texture_Correlation_ER_3_0', 'Cytoplasm_Texture_Correlation_ER_5_0', 'Cytoplasm_Texture_Correlation_Mito_10_0', 'Cytoplasm_Texture_Correlation_Mito_3_0', 'Cytoplasm_Texture_Correlation_Mito_5_0', 'Cytoplasm_Texture_Correlation_RNA_10_0', 'Cytoplasm_Texture_Correlation_RNA_3_0', 'Cytoplasm_Texture_Correlation_RNA_5_0', 'Cytoplasm_Texture_DifferenceEntropy_AGP_10_0', 'Cytoplasm_Texture_DifferenceEntropy_AGP_3_0', 'Cytoplasm_Texture_DifferenceEntropy_AGP_5_0', 'Cytoplasm_Texture_DifferenceEntropy_DNA_10_0', 'Cytoplasm_Texture_DifferenceEntropy_DNA_3_0', 'Cytoplasm_Texture_DifferenceEntropy_DNA_5_0', 'Cytoplasm_Texture_DifferenceEntropy_ER_10_0', 'Cytoplasm_Texture_DifferenceEntropy_ER_3_0', 'Cytoplasm_Texture_DifferenceEntropy_ER_5_0', 'Cytoplasm_Texture_DifferenceEntropy_Mito_10_0', 'Cytoplasm_Texture_DifferenceEntropy_Mito_3_0', 'Cytoplasm_Texture_DifferenceEntropy_Mito_5_0', 'Cytoplasm_Texture_DifferenceEntropy_RNA_10_0', 'Cytoplasm_Texture_DifferenceEntropy_RNA_3_0', 'Cytoplasm_Texture_DifferenceEntropy_RNA_5_0', 'Cytoplasm_Texture_DifferenceVariance_AGP_10_0', 'Cytoplasm_Texture_DifferenceVariance_AGP_3_0', 'Cytoplasm_Texture_DifferenceVariance_AGP_5_0', 'Cytoplasm_Texture_DifferenceVariance_DNA_10_0', 'Cytoplasm_Texture_DifferenceVariance_DNA_3_0', 'Cytoplasm_Texture_DifferenceVariance_DNA_5_0', 'Cytoplasm_Texture_DifferenceVariance_ER_10_0', 'Cytoplasm_Texture_DifferenceVariance_ER_3_0', 'Cytoplasm_Texture_DifferenceVariance_ER_5_0', 'Cytoplasm_Texture_DifferenceVariance_Mito_10_0', 'Cytoplasm_Texture_DifferenceVariance_Mito_3_0', 'Cytoplasm_Texture_DifferenceVariance_Mito_5_0', 'Cytoplasm_Texture_DifferenceVariance_RNA_10_0', 'Cytoplasm_Texture_DifferenceVariance_RNA_3_0', 'Cytoplasm_Texture_DifferenceVariance_RNA_5_0', 'Cytoplasm_Texture_Entropy_AGP_10_0', 'Cytoplasm_Texture_Entropy_AGP_3_0', 'Cytoplasm_Texture_Entropy_AGP_5_0', 'Cytoplasm_Texture_Entropy_DNA_10_0', 'Cytoplasm_Texture_Entropy_DNA_3_0', 'Cytoplasm_Texture_Entropy_DNA_5_0', 'Cytoplasm_Texture_Entropy_ER_10_0', 'Cytoplasm_Texture_Entropy_ER_3_0', 'Cytoplasm_Texture_Entropy_ER_5_0', 'Cytoplasm_Texture_Entropy_Mito_10_0', 'Cytoplasm_Texture_Entropy_Mito_3_0', 'Cytoplasm_Texture_Entropy_Mito_5_0', 'Cytoplasm_Texture_Entropy_RNA_10_0', 'Cytoplasm_Texture_Entropy_RNA_3_0', 'Cytoplasm_Texture_Entropy_RNA_5_0', 'Cytoplasm_Texture_Gabor_AGP_10', 'Cytoplasm_Texture_Gabor_AGP_3', 'Cytoplasm_Texture_Gabor_AGP_5', 'Cytoplasm_Texture_Gabor_DNA_10', 'Cytoplasm_Texture_Gabor_DNA_3', 'Cytoplasm_Texture_Gabor_DNA_5', 'Cytoplasm_Texture_Gabor_ER_10', 'Cytoplasm_Texture_Gabor_ER_3', 'Cytoplasm_Texture_Gabor_ER_5', 'Cytoplasm_Texture_Gabor_Mito_10', 'Cytoplasm_Texture_Gabor_Mito_3', 'Cytoplasm_Texture_Gabor_Mito_5', 'Cytoplasm_Texture_Gabor_RNA_10', 'Cytoplasm_Texture_Gabor_RNA_3', 'Cytoplasm_Texture_Gabor_RNA_5', 'Cytoplasm_Texture_InfoMeas1_AGP_10_0', 'Cytoplasm_Texture_InfoMeas1_AGP_3_0', 'Cytoplasm_Texture_InfoMeas1_AGP_5_0', 'Cytoplasm_Texture_InfoMeas1_DNA_10_0', 'Cytoplasm_Texture_InfoMeas1_DNA_3_0', 'Cytoplasm_Texture_InfoMeas1_DNA_5_0', 'Cytoplasm_Texture_InfoMeas1_ER_10_0', 'Cytoplasm_Texture_InfoMeas1_ER_3_0', 'Cytoplasm_Texture_InfoMeas1_ER_5_0', 'Cytoplasm_Texture_InfoMeas1_Mito_10_0', 'Cytoplasm_Texture_InfoMeas1_Mito_3_0', 'Cytoplasm_Texture_InfoMeas1_Mito_5_0', 'Cytoplasm_Texture_InfoMeas1_RNA_10_0', 'Cytoplasm_Texture_InfoMeas1_RNA_3_0', 'Cytoplasm_Texture_InfoMeas1_RNA_5_0', 'Cytoplasm_Texture_InfoMeas2_AGP_10_0', 'Cytoplasm_Texture_InfoMeas2_AGP_3_0', 'Cytoplasm_Texture_InfoMeas2_AGP_5_0', 'Cytoplasm_Texture_InfoMeas2_DNA_10_0', 'Cytoplasm_Texture_InfoMeas2_DNA_3_0', 'Cytoplasm_Texture_InfoMeas2_DNA_5_0', 'Cytoplasm_Texture_InfoMeas2_ER_10_0', 'Cytoplasm_Texture_InfoMeas2_ER_3_0', 'Cytoplasm_Texture_InfoMeas2_ER_5_0', 'Cytoplasm_Texture_InfoMeas2_Mito_10_0', 'Cytoplasm_Texture_InfoMeas2_Mito_3_0', 'Cytoplasm_Texture_InfoMeas2_Mito_5_0', 'Cytoplasm_Texture_InfoMeas2_RNA_10_0', 'Cytoplasm_Texture_InfoMeas2_RNA_3_0', 'Cytoplasm_Texture_InfoMeas2_RNA_5_0', 'Cytoplasm_Texture_InverseDifferenceMoment_AGP_10_0', 'Cytoplasm_Texture_InverseDifferenceMoment_AGP_3_0', 'Cytoplasm_Texture_InverseDifferenceMoment_AGP_5_0', 'Cytoplasm_Texture_InverseDifferenceMoment_DNA_10_0', 'Cytoplasm_Texture_InverseDifferenceMoment_DNA_3_0', 'Cytoplasm_Texture_InverseDifferenceMoment_DNA_5_0', 'Cytoplasm_Texture_InverseDifferenceMoment_ER_10_0', 'Cytoplasm_Texture_InverseDifferenceMoment_ER_3_0', 'Cytoplasm_Texture_InverseDifferenceMoment_ER_5_0', 'Cytoplasm_Texture_InverseDifferenceMoment_Mito_10_0', 'Cytoplasm_Texture_InverseDifferenceMoment_Mito_3_0', 'Cytoplasm_Texture_InverseDifferenceMoment_Mito_5_0', 'Cytoplasm_Texture_InverseDifferenceMoment_RNA_10_0', 'Cytoplasm_Texture_InverseDifferenceMoment_RNA_3_0', 'Cytoplasm_Texture_InverseDifferenceMoment_RNA_5_0', 'Cytoplasm_Texture_SumAverage_AGP_10_0', 'Cytoplasm_Texture_SumAverage_AGP_3_0', 'Cytoplasm_Texture_SumAverage_AGP_5_0', 'Cytoplasm_Texture_SumAverage_DNA_10_0', 'Cytoplasm_Texture_SumAverage_DNA_3_0', 'Cytoplasm_Texture_SumAverage_DNA_5_0', 'Cytoplasm_Texture_SumAverage_ER_10_0', 'Cytoplasm_Texture_SumAverage_ER_3_0', 'Cytoplasm_Texture_SumAverage_ER_5_0', 'Cytoplasm_Texture_SumAverage_Mito_10_0', 'Cytoplasm_Texture_SumAverage_Mito_3_0', 'Cytoplasm_Texture_SumAverage_Mito_5_0', 'Cytoplasm_Texture_SumAverage_RNA_10_0', 'Cytoplasm_Texture_SumAverage_RNA_3_0', 'Cytoplasm_Texture_SumAverage_RNA_5_0', 'Cytoplasm_Texture_SumEntropy_AGP_10_0', 'Cytoplasm_Texture_SumEntropy_AGP_3_0', 'Cytoplasm_Texture_SumEntropy_AGP_5_0', 'Cytoplasm_Texture_SumEntropy_DNA_10_0', 'Cytoplasm_Texture_SumEntropy_DNA_3_0', 'Cytoplasm_Texture_SumEntropy_DNA_5_0', 'Cytoplasm_Texture_SumEntropy_ER_10_0', 'Cytoplasm_Texture_SumEntropy_ER_3_0', 'Cytoplasm_Texture_SumEntropy_ER_5_0', 'Cytoplasm_Texture_SumEntropy_Mito_10_0', 'Cytoplasm_Texture_SumEntropy_Mito_3_0', 'Cytoplasm_Texture_SumEntropy_Mito_5_0', 'Cytoplasm_Texture_SumEntropy_RNA_10_0', 'Cytoplasm_Texture_SumEntropy_RNA_3_0', 'Cytoplasm_Texture_SumEntropy_RNA_5_0', 'Cytoplasm_Texture_SumVariance_AGP_10_0', 'Cytoplasm_Texture_SumVariance_AGP_3_0', 'Cytoplasm_Texture_SumVariance_AGP_5_0', 'Cytoplasm_Texture_SumVariance_DNA_10_0', 'Cytoplasm_Texture_SumVariance_DNA_3_0', 'Cytoplasm_Texture_SumVariance_DNA_5_0', 'Cytoplasm_Texture_SumVariance_ER_10_0', 'Cytoplasm_Texture_SumVariance_ER_3_0', 'Cytoplasm_Texture_SumVariance_ER_5_0', 'Cytoplasm_Texture_SumVariance_Mito_10_0', 'Cytoplasm_Texture_SumVariance_Mito_3_0', 'Cytoplasm_Texture_SumVariance_Mito_5_0', 'Cytoplasm_Texture_SumVariance_RNA_10_0', 'Cytoplasm_Texture_SumVariance_RNA_3_0', 'Cytoplasm_Texture_SumVariance_RNA_5_0', 'Cytoplasm_Texture_Variance_AGP_10_0', 'Cytoplasm_Texture_Variance_AGP_3_0', 'Cytoplasm_Texture_Variance_AGP_5_0', 'Cytoplasm_Texture_Variance_DNA_10_0', 'Cytoplasm_Texture_Variance_DNA_3_0', 'Cytoplasm_Texture_Variance_DNA_5_0', 'Cytoplasm_Texture_Variance_ER_10_0', 'Cytoplasm_Texture_Variance_ER_3_0', 'Cytoplasm_Texture_Variance_ER_5_0', 'Cytoplasm_Texture_Variance_Mito_10_0', 'Cytoplasm_Texture_Variance_Mito_3_0', 'Cytoplasm_Texture_Variance_Mito_5_0', 'Cytoplasm_Texture_Variance_RNA_10_0', 'Cytoplasm_Texture_Variance_RNA_3_0', 'Cytoplasm_Texture_Variance_RNA_5_0'

Nuclei Table

'Nuclei_AreaShape_Area', 'Nuclei_AreaShape_Center_X', 'Nuclei_AreaShape_Center_Y', 'Nuclei_AreaShape_Compactness', 'Nuclei_AreaShape_Eccentricity', 'Nuclei_AreaShape_EulerNumber', 'Nuclei_AreaShape_Extent', 'Nuclei_AreaShape_FormFactor', 'Nuclei_AreaShape_MajorAxisLength', 'Nuclei_AreaShape_MaxFeretDiameter', 'Nuclei_AreaShape_MaximumRadius', 'Nuclei_AreaShape_MeanRadius', 'Nuclei_AreaShape_MedianRadius', 'Nuclei_AreaShape_MinFeretDiameter', 'Nuclei_AreaShape_MinorAxisLength', 'Nuclei_AreaShape_Orientation', 'Nuclei_AreaShape_Perimeter', 'Nuclei_AreaShape_Solidity', 'Nuclei_AreaShape_Zernike_0_0', 'Nuclei_AreaShape_Zernike_1_1', 'Nuclei_AreaShape_Zernike_2_0', 'Nuclei_AreaShape_Zernike_2_2', 'Nuclei_AreaShape_Zernike_3_1', 'Nuclei_AreaShape_Zernike_3_3', 'Nuclei_AreaShape_Zernike_4_0', 'Nuclei_AreaShape_Zernike_4_2', 'Nuclei_AreaShape_Zernike_4_4', 'Nuclei_AreaShape_Zernike_5_1', 'Nuclei_AreaShape_Zernike_5_3', 'Nuclei_AreaShape_Zernike_5_5', 'Nuclei_AreaShape_Zernike_6_0', 'Nuclei_AreaShape_Zernike_6_2', 'Nuclei_AreaShape_Zernike_6_4', 'Nuclei_AreaShape_Zernike_6_6', 'Nuclei_AreaShape_Zernike_7_1', 'Nuclei_AreaShape_Zernike_7_3', 'Nuclei_AreaShape_Zernike_7_5', 'Nuclei_AreaShape_Zernike_7_7', 'Nuclei_AreaShape_Zernike_8_0', 'Nuclei_AreaShape_Zernike_8_2', 'Nuclei_AreaShape_Zernike_8_4', 'Nuclei_AreaShape_Zernike_8_6', 'Nuclei_AreaShape_Zernike_8_8', 'Nuclei_AreaShape_Zernike_9_1', 'Nuclei_AreaShape_Zernike_9_3', 'Nuclei_AreaShape_Zernike_9_5', 'Nuclei_AreaShape_Zernike_9_7', 'Nuclei_AreaShape_Zernike_9_9', 'Nuclei_Children_Cells_Count', 'Nuclei_Children_Cytoplasm_Count', 'Nuclei_Correlation_Correlation_DNA_AGP', 'Nuclei_Correlation_Correlation_DNA_ER', 'Nuclei_Correlation_Correlation_DNA_Mito', 'Nuclei_Correlation_Correlation_DNA_RNA', 'Nuclei_Correlation_Correlation_ER_AGP', 'Nuclei_Correlation_Correlation_ER_RNA', 'Nuclei_Correlation_Correlation_Mito_AGP', 'Nuclei_Correlation_Correlation_Mito_ER', 'Nuclei_Correlation_Correlation_Mito_RNA', 'Nuclei_Correlation_Correlation_RNA_AGP', 'Nuclei_Correlation_Costes_AGP_DNA', 'Nuclei_Correlation_Costes_AGP_ER', 'Nuclei_Correlation_Costes_AGP_Mito', 'Nuclei_Correlation_Costes_AGP_RNA', 'Nuclei_Correlation_Costes_DNA_AGP', 'Nuclei_Correlation_Costes_DNA_ER', 'Nuclei_Correlation_Costes_DNA_Mito', 'Nuclei_Correlation_Costes_DNA_RNA', 'Nuclei_Correlation_Costes_ER_AGP', 'Nuclei_Correlation_Costes_ER_DNA', 'Nuclei_Correlation_Costes_ER_Mito', 'Nuclei_Correlation_Costes_ER_RNA', 'Nuclei_Correlation_Costes_Mito_AGP', 'Nuclei_Correlation_Costes_Mito_DNA', 'Nuclei_Correlation_Costes_Mito_ER', 'Nuclei_Correlation_Costes_Mito_RNA', 'Nuclei_Correlation_Costes_RNA_AGP', 'Nuclei_Correlation_Costes_RNA_DNA', 'Nuclei_Correlation_Costes_RNA_ER', 'Nuclei_Correlation_Costes_RNA_Mito', 'Nuclei_Correlation_K_AGP_DNA', 'Nuclei_Correlation_K_AGP_ER', 'Nuclei_Correlation_K_AGP_Mito', 'Nuclei_Correlation_K_AGP_RNA', 'Nuclei_Correlation_K_DNA_AGP', 'Nuclei_Correlation_K_DNA_ER', 'Nuclei_Correlation_K_DNA_Mito', 'Nuclei_Correlation_K_DNA_RNA', 'Nuclei_Correlation_K_ER_AGP', 'Nuclei_Correlation_K_ER_DNA', 'Nuclei_Correlation_K_ER_Mito', 'Nuclei_Correlation_K_ER_RNA', 'Nuclei_Correlation_K_Mito_AGP', 'Nuclei_Correlation_K_Mito_DNA', 'Nuclei_Correlation_K_Mito_ER', 'Nuclei_Correlation_K_Mito_RNA', 'Nuclei_Correlation_K_RNA_AGP', 'Nuclei_Correlation_K_RNA_DNA', 'Nuclei_Correlation_K_RNA_ER', 'Nuclei_Correlation_K_RNA_Mito', 'Nuclei_Correlation_Manders_AGP_DNA', 'Nuclei_Correlation_Manders_AGP_ER', 'Nuclei_Correlation_Manders_AGP_Mito', 'Nuclei_Correlation_Manders_AGP_RNA', 'Nuclei_Correlation_Manders_DNA_AGP', 'Nuclei_Correlation_Manders_DNA_ER', 'Nuclei_Correlation_Manders_DNA_Mito', 'Nuclei_Correlation_Manders_DNA_RNA', 'Nuclei_Correlation_Manders_ER_AGP', 'Nuclei_Correlation_Manders_ER_DNA', 'Nuclei_Correlation_Manders_ER_Mito', 'Nuclei_Correlation_Manders_ER_RNA', 'Nuclei_Correlation_Manders_Mito_AGP', 'Nuclei_Correlation_Manders_Mito_DNA', 'Nuclei_Correlation_Manders_Mito_ER', 'Nuclei_Correlation_Manders_Mito_RNA', 'Nuclei_Correlation_Manders_RNA_AGP', 'Nuclei_Correlation_Manders_RNA_DNA', 'Nuclei_Correlation_Manders_RNA_ER', 'Nuclei_Correlation_Manders_RNA_Mito', 'Nuclei_Correlation_Overlap_DNA_AGP', 'Nuclei_Correlation_Overlap_DNA_ER', 'Nuclei_Correlation_Overlap_DNA_Mito', 'Nuclei_Correlation_Overlap_DNA_RNA', 'Nuclei_Correlation_Overlap_ER_AGP', 'Nuclei_Correlation_Overlap_ER_RNA', 'Nuclei_Correlation_Overlap_Mito_AGP', 'Nuclei_Correlation_Overlap_Mito_ER', 'Nuclei_Correlation_Overlap_Mito_RNA', 'Nuclei_Correlation_Overlap_RNA_AGP', 'Nuclei_Correlation_RWC_AGP_DNA', 'Nuclei_Correlation_RWC_AGP_ER', 'Nuclei_Correlation_RWC_AGP_Mito', 'Nuclei_Correlation_RWC_AGP_RNA', 'Nuclei_Correlation_RWC_DNA_AGP', 'Nuclei_Correlation_RWC_DNA_ER', 'Nuclei_Correlation_RWC_DNA_Mito', 'Nuclei_Correlation_RWC_DNA_RNA', 'Nuclei_Correlation_RWC_ER_AGP', 'Nuclei_Correlation_RWC_ER_DNA', 'Nuclei_Correlation_RWC_ER_Mito', 'Nuclei_Correlation_RWC_ER_RNA', 'Nuclei_Correlation_RWC_Mito_AGP', 'Nuclei_Correlation_RWC_Mito_DNA', 'Nuclei_Correlation_RWC_Mito_ER', 'Nuclei_Correlation_RWC_Mito_RNA', 'Nuclei_Correlation_RWC_RNA_AGP', 'Nuclei_Correlation_RWC_RNA_DNA', 'Nuclei_Correlation_RWC_RNA_ER', 'Nuclei_Correlation_RWC_RNA_Mito', 'Nuclei_Granularity_10_AGP', 'Nuclei_Granularity_10_DNA', 'Nuclei_Granularity_10_ER', 'Nuclei_Granularity_10_Mito', 'Nuclei_Granularity_10_RNA', 'Nuclei_Granularity_11_AGP', 'Nuclei_Granularity_11_DNA', 'Nuclei_Granularity_11_ER', 'Nuclei_Granularity_11_Mito', 'Nuclei_Granularity_11_RNA', 'Nuclei_Granularity_12_AGP', 'Nuclei_Granularity_12_DNA', 'Nuclei_Granularity_12_ER', 'Nuclei_Granularity_12_Mito', 'Nuclei_Granularity_12_RNA', 'Nuclei_Granularity_13_AGP', 'Nuclei_Granularity_13_DNA', 'Nuclei_Granularity_13_ER', 'Nuclei_Granularity_13_Mito', 'Nuclei_Granularity_13_RNA', 'Nuclei_Granularity_14_AGP', 'Nuclei_Granularity_14_DNA', 'Nuclei_Granularity_14_ER', 'Nuclei_Granularity_14_Mito', 'Nuclei_Granularity_14_RNA', 'Nuclei_Granularity_15_AGP', 'Nuclei_Granularity_15_DNA', 'Nuclei_Granularity_15_ER', 'Nuclei_Granularity_15_Mito', 'Nuclei_Granularity_15_RNA', 'Nuclei_Granularity_16_AGP', 'Nuclei_Granularity_16_DNA', 'Nuclei_Granularity_16_ER', 'Nuclei_Granularity_16_Mito', 'Nuclei_Granularity_16_RNA', 'Nuclei_Granularity_1_AGP', 'Nuclei_Granularity_1_DNA', 'Nuclei_Granularity_1_ER', 'Nuclei_Granularity_1_Mito', 'Nuclei_Granularity_1_RNA', 'Nuclei_Granularity_2_AGP', 'Nuclei_Granularity_2_DNA', 'Nuclei_Granularity_2_ER', 'Nuclei_Granularity_2_Mito', 'Nuclei_Granularity_2_RNA', 'Nuclei_Granularity_3_AGP', 'Nuclei_Granularity_3_DNA', 'Nuclei_Granularity_3_ER', 'Nuclei_Granularity_3_Mito', 'Nuclei_Granularity_3_RNA', 'Nuclei_Granularity_4_AGP', 'Nuclei_Granularity_4_DNA', 'Nuclei_Granularity_4_ER', 'Nuclei_Granularity_4_Mito', 'Nuclei_Granularity_4_RNA', 'Nuclei_Granularity_5_AGP', 'Nuclei_Granularity_5_DNA', 'Nuclei_Granularity_5_ER', 'Nuclei_Granularity_5_Mito', 'Nuclei_Granularity_5_RNA', 'Nuclei_Granularity_6_AGP', 'Nuclei_Granularity_6_DNA', 'Nuclei_Granularity_6_ER', 'Nuclei_Granularity_6_Mito', 'Nuclei_Granularity_6_RNA', 'Nuclei_Granularity_7_AGP', 'Nuclei_Granularity_7_DNA', 'Nuclei_Granularity_7_ER', 'Nuclei_Granularity_7_Mito', 'Nuclei_Granularity_7_RNA', 'Nuclei_Granularity_8_AGP', 'Nuclei_Granularity_8_DNA', 'Nuclei_Granularity_8_ER', 'Nuclei_Granularity_8_Mito', 'Nuclei_Granularity_8_RNA', 'Nuclei_Granularity_9_AGP', 'Nuclei_Granularity_9_DNA', 'Nuclei_Granularity_9_ER', 'Nuclei_Granularity_9_Mito', 'Nuclei_Granularity_9_RNA', 'Nuclei_Intensity_IntegratedIntensityEdge_AGP', 'Nuclei_Intensity_IntegratedIntensityEdge_DNA', 'Nuclei_Intensity_IntegratedIntensityEdge_ER', 'Nuclei_Intensity_IntegratedIntensityEdge_Mito', 'Nuclei_Intensity_IntegratedIntensityEdge_RNA', 'Nuclei_Intensity_IntegratedIntensity_AGP', 'Nuclei_Intensity_IntegratedIntensity_DNA', 'Nuclei_Intensity_IntegratedIntensity_ER', 'Nuclei_Intensity_IntegratedIntensity_Mito', 'Nuclei_Intensity_IntegratedIntensity_RNA', 'Nuclei_Intensity_LowerQuartileIntensity_AGP', 'Nuclei_Intensity_LowerQuartileIntensity_DNA', 'Nuclei_Intensity_LowerQuartileIntensity_ER', 'Nuclei_Intensity_LowerQuartileIntensity_Mito', 'Nuclei_Intensity_LowerQuartileIntensity_RNA', 'Nuclei_Intensity_MADIntensity_AGP', 'Nuclei_Intensity_MADIntensity_DNA', 'Nuclei_Intensity_MADIntensity_ER', 'Nuclei_Intensity_MADIntensity_Mito', 'Nuclei_Intensity_MADIntensity_RNA', 'Nuclei_Intensity_MassDisplacement_AGP', 'Nuclei_Intensity_MassDisplacement_DNA', 'Nuclei_Intensity_MassDisplacement_ER', 'Nuclei_Intensity_MassDisplacement_Mito', 'Nuclei_Intensity_MassDisplacement_RNA', 'Nuclei_Intensity_MaxIntensityEdge_AGP', 'Nuclei_Intensity_MaxIntensityEdge_DNA', 'Nuclei_Intensity_MaxIntensityEdge_ER', 'Nuclei_Intensity_MaxIntensityEdge_Mito', 'Nuclei_Intensity_MaxIntensityEdge_RNA', 'Nuclei_Intensity_MaxIntensity_AGP', 'Nuclei_Intensity_MaxIntensity_DNA', 'Nuclei_Intensity_MaxIntensity_ER', 'Nuclei_Intensity_MaxIntensity_Mito', 'Nuclei_Intensity_MaxIntensity_RNA', 'Nuclei_Intensity_MeanIntensityEdge_AGP', 'Nuclei_Intensity_MeanIntensityEdge_DNA', 'Nuclei_Intensity_MeanIntensityEdge_ER', 'Nuclei_Intensity_MeanIntensityEdge_Mito', 'Nuclei_Intensity_MeanIntensityEdge_RNA', 'Nuclei_Intensity_MeanIntensity_AGP', 'Nuclei_Intensity_MeanIntensity_DNA', 'Nuclei_Intensity_MeanIntensity_ER', 'Nuclei_Intensity_MeanIntensity_Mito', 'Nuclei_Intensity_MeanIntensity_RNA', 'Nuclei_Intensity_MedianIntensity_AGP', 'Nuclei_Intensity_MedianIntensity_DNA', 'Nuclei_Intensity_MedianIntensity_ER', 'Nuclei_Intensity_MedianIntensity_Mito', 'Nuclei_Intensity_MedianIntensity_RNA', 'Nuclei_Intensity_MinIntensityEdge_AGP', 'Nuclei_Intensity_MinIntensityEdge_DNA', 'Nuclei_Intensity_MinIntensityEdge_ER', 'Nuclei_Intensity_MinIntensityEdge_Mito', 'Nuclei_Intensity_MinIntensityEdge_RNA', 'Nuclei_Intensity_MinIntensity_AGP', 'Nuclei_Intensity_MinIntensity_DNA', 'Nuclei_Intensity_MinIntensity_ER', 'Nuclei_Intensity_MinIntensity_Mito', 'Nuclei_Intensity_MinIntensity_RNA', 'Nuclei_Intensity_StdIntensityEdge_AGP', 'Nuclei_Intensity_StdIntensityEdge_DNA', 'Nuclei_Intensity_StdIntensityEdge_ER', 'Nuclei_Intensity_StdIntensityEdge_Mito', 'Nuclei_Intensity_StdIntensityEdge_RNA', 'Nuclei_Intensity_StdIntensity_AGP', 'Nuclei_Intensity_StdIntensity_DNA', 'Nuclei_Intensity_StdIntensity_ER', 'Nuclei_Intensity_StdIntensity_Mito', 'Nuclei_Intensity_StdIntensity_RNA', 'Nuclei_Intensity_UpperQuartileIntensity_AGP', 'Nuclei_Intensity_UpperQuartileIntensity_DNA', 'Nuclei_Intensity_UpperQuartileIntensity_ER', 'Nuclei_Intensity_UpperQuartileIntensity_Mito', 'Nuclei_Intensity_UpperQuartileIntensity_RNA', 'Nuclei_Location_CenterMassIntensity_X_AGP', 'Nuclei_Location_CenterMassIntensity_X_DNA', 'Nuclei_Location_CenterMassIntensity_X_ER', 'Nuclei_Location_CenterMassIntensity_X_Mito', 'Nuclei_Location_CenterMassIntensity_X_RNA', 'Nuclei_Location_CenterMassIntensity_Y_AGP', 'Nuclei_Location_CenterMassIntensity_Y_DNA', 'Nuclei_Location_CenterMassIntensity_Y_ER', 'Nuclei_Location_CenterMassIntensity_Y_Mito', 'Nuclei_Location_CenterMassIntensity_Y_RNA', 'Nuclei_Location_Center_X', 'Nuclei_Location_Center_Y', 'Nuclei_Location_MaxIntensity_X_AGP', 'Nuclei_Location_MaxIntensity_X_DNA', 'Nuclei_Location_MaxIntensity_X_ER', 'Nuclei_Location_MaxIntensity_X_Mito', 'Nuclei_Location_MaxIntensity_X_RNA', 'Nuclei_Location_MaxIntensity_Y_AGP', 'Nuclei_Location_MaxIntensity_Y_DNA', 'Nuclei_Location_MaxIntensity_Y_ER', 'Nuclei_Location_MaxIntensity_Y_Mito', 'Nuclei_Location_MaxIntensity_Y_RNA', 'Nuclei_Neighbors_AngleBetweenNeighbors_1', 'Nuclei_Neighbors_FirstClosestDistance_1', 'Nuclei_Neighbors_FirstClosestObjectNumber_1', 'Nuclei_Neighbors_NumberOfNeighbors_1', 'Nuclei_Neighbors_PercentTouching_1', 'Nuclei_Neighbors_SecondClosestDistance_1', 'Nuclei_Neighbors_SecondClosestObjectNumber_1', 'Nuclei_Number_Object_Number', 'Nuclei_RadialDistribution_FracAtD_AGP_1of4', 'Nuclei_RadialDistribution_FracAtD_AGP_2of4', 'Nuclei_RadialDistribution_FracAtD_AGP_3of4', 'Nuclei_RadialDistribution_FracAtD_AGP_4of4', 'Nuclei_RadialDistribution_FracAtD_DNA_1of4', 'Nuclei_RadialDistribution_FracAtD_DNA_2of4', 'Nuclei_RadialDistribution_FracAtD_DNA_3of4', 'Nuclei_RadialDistribution_FracAtD_DNA_4of4', 'Nuclei_RadialDistribution_FracAtD_ER_1of4', 'Nuclei_RadialDistribution_FracAtD_ER_2of4', 'Nuclei_RadialDistribution_FracAtD_ER_3of4', 'Nuclei_RadialDistribution_FracAtD_ER_4of4', 'Nuclei_RadialDistribution_FracAtD_Mito_1of4', 'Nuclei_RadialDistribution_FracAtD_Mito_2of4', 'Nuclei_RadialDistribution_FracAtD_Mito_3of4', 'Nuclei_RadialDistribution_FracAtD_Mito_4of4', 'Nuclei_RadialDistribution_FracAtD_RNA_1of4', 'Nuclei_RadialDistribution_FracAtD_RNA_2of4', 'Nuclei_RadialDistribution_FracAtD_RNA_3of4', 'Nuclei_RadialDistribution_FracAtD_RNA_4of4', 'Nuclei_RadialDistribution_MeanFrac_AGP_1of4', 'Nuclei_RadialDistribution_MeanFrac_AGP_2of4', 'Nuclei_RadialDistribution_MeanFrac_AGP_3of4', 'Nuclei_RadialDistribution_MeanFrac_AGP_4of4', 'Nuclei_RadialDistribution_MeanFrac_DNA_1of4', 'Nuclei_RadialDistribution_MeanFrac_DNA_2of4', 'Nuclei_RadialDistribution_MeanFrac_DNA_3of4', 'Nuclei_RadialDistribution_MeanFrac_DNA_4of4', 'Nuclei_RadialDistribution_MeanFrac_ER_1of4', 'Nuclei_RadialDistribution_MeanFrac_ER_2of4', 'Nuclei_RadialDistribution_MeanFrac_ER_3of4', 'Nuclei_RadialDistribution_MeanFrac_ER_4of4', 'Nuclei_RadialDistribution_MeanFrac_Mito_1of4', 'Nuclei_RadialDistribution_MeanFrac_Mito_2of4', 'Nuclei_RadialDistribution_MeanFrac_Mito_3of4', 'Nuclei_RadialDistribution_MeanFrac_Mito_4of4', 'Nuclei_RadialDistribution_MeanFrac_RNA_1of4', 'Nuclei_RadialDistribution_MeanFrac_RNA_2of4', 'Nuclei_RadialDistribution_MeanFrac_RNA_3of4', 'Nuclei_RadialDistribution_MeanFrac_RNA_4of4', 'Nuclei_RadialDistribution_RadialCV_AGP_1of4', 'Nuclei_RadialDistribution_RadialCV_AGP_2of4', 'Nuclei_RadialDistribution_RadialCV_AGP_3of4', 'Nuclei_RadialDistribution_RadialCV_AGP_4of4', 'Nuclei_RadialDistribution_RadialCV_DNA_1of4', 'Nuclei_RadialDistribution_RadialCV_DNA_2of4', 'Nuclei_RadialDistribution_RadialCV_DNA_3of4', 'Nuclei_RadialDistribution_RadialCV_DNA_4of4', 'Nuclei_RadialDistribution_RadialCV_ER_1of4', 'Nuclei_RadialDistribution_RadialCV_ER_2of4', 'Nuclei_RadialDistribution_RadialCV_ER_3of4', 'Nuclei_RadialDistribution_RadialCV_ER_4of4', 'Nuclei_RadialDistribution_RadialCV_Mito_1of4', 'Nuclei_RadialDistribution_RadialCV_Mito_2of4', 'Nuclei_RadialDistribution_RadialCV_Mito_3of4', 'Nuclei_RadialDistribution_RadialCV_Mito_4of4', 'Nuclei_RadialDistribution_RadialCV_RNA_1of4', 'Nuclei_RadialDistribution_RadialCV_RNA_2of4', 'Nuclei_RadialDistribution_RadialCV_RNA_3of4', 'Nuclei_RadialDistribution_RadialCV_RNA_4of4', 'Nuclei_Texture_AngularSecondMoment_AGP_10_0', 'Nuclei_Texture_AngularSecondMoment_AGP_3_0', 'Nuclei_Texture_AngularSecondMoment_AGP_5_0', 'Nuclei_Texture_AngularSecondMoment_DNA_10_0', 'Nuclei_Texture_AngularSecondMoment_DNA_3_0', 'Nuclei_Texture_AngularSecondMoment_DNA_5_0', 'Nuclei_Texture_AngularSecondMoment_ER_10_0', 'Nuclei_Texture_AngularSecondMoment_ER_3_0', 'Nuclei_Texture_AngularSecondMoment_ER_5_0', 'Nuclei_Texture_AngularSecondMoment_Mito_10_0', 'Nuclei_Texture_AngularSecondMoment_Mito_3_0', 'Nuclei_Texture_AngularSecondMoment_Mito_5_0', 'Nuclei_Texture_AngularSecondMoment_RNA_10_0', 'Nuclei_Texture_AngularSecondMoment_RNA_3_0', 'Nuclei_Texture_AngularSecondMoment_RNA_5_0', 'Nuclei_Texture_Contrast_AGP_10_0', 'Nuclei_Texture_Contrast_AGP_3_0', 'Nuclei_Texture_Contrast_AGP_5_0', 'Nuclei_Texture_Contrast_DNA_10_0', 'Nuclei_Texture_Contrast_DNA_3_0', 'Nuclei_Texture_Contrast_DNA_5_0', 'Nuclei_Texture_Contrast_ER_10_0', 'Nuclei_Texture_Contrast_ER_3_0', 'Nuclei_Texture_Contrast_ER_5_0', 'Nuclei_Texture_Contrast_Mito_10_0', 'Nuclei_Texture_Contrast_Mito_3_0', 'Nuclei_Texture_Contrast_Mito_5_0', 'Nuclei_Texture_Contrast_RNA_10_0', 'Nuclei_Texture_Contrast_RNA_3_0', 'Nuclei_Texture_Contrast_RNA_5_0', 'Nuclei_Texture_Correlation_AGP_10_0', 'Nuclei_Texture_Correlation_AGP_3_0', 'Nuclei_Texture_Correlation_AGP_5_0', 'Nuclei_Texture_Correlation_DNA_10_0', 'Nuclei_Texture_Correlation_DNA_3_0', 'Nuclei_Texture_Correlation_DNA_5_0', 'Nuclei_Texture_Correlation_ER_10_0', 'Nuclei_Texture_Correlation_ER_3_0', 'Nuclei_Texture_Correlation_ER_5_0', 'Nuclei_Texture_Correlation_Mito_10_0', 'Nuclei_Texture_Correlation_Mito_3_0', 'Nuclei_Texture_Correlation_Mito_5_0', 'Nuclei_Texture_Correlation_RNA_10_0', 'Nuclei_Texture_Correlation_RNA_3_0', 'Nuclei_Texture_Correlation_RNA_5_0', 'Nuclei_Texture_DifferenceEntropy_AGP_10_0', 'Nuclei_Texture_DifferenceEntropy_AGP_3_0', 'Nuclei_Texture_DifferenceEntropy_AGP_5_0', 'Nuclei_Texture_DifferenceEntropy_DNA_10_0', 'Nuclei_Texture_DifferenceEntropy_DNA_3_0', 'Nuclei_Texture_DifferenceEntropy_DNA_5_0', 'Nuclei_Texture_DifferenceEntropy_ER_10_0', 'Nuclei_Texture_DifferenceEntropy_ER_3_0', 'Nuclei_Texture_DifferenceEntropy_ER_5_0', 'Nuclei_Texture_DifferenceEntropy_Mito_10_0', 'Nuclei_Texture_DifferenceEntropy_Mito_3_0', 'Nuclei_Texture_DifferenceEntropy_Mito_5_0', 'Nuclei_Texture_DifferenceEntropy_RNA_10_0', 'Nuclei_Texture_DifferenceEntropy_RNA_3_0', 'Nuclei_Texture_DifferenceEntropy_RNA_5_0', 'Nuclei_Texture_DifferenceVariance_AGP_10_0', 'Nuclei_Texture_DifferenceVariance_AGP_3_0', 'Nuclei_Texture_DifferenceVariance_AGP_5_0', 'Nuclei_Texture_DifferenceVariance_DNA_10_0', 'Nuclei_Texture_DifferenceVariance_DNA_3_0', 'Nuclei_Texture_DifferenceVariance_DNA_5_0', 'Nuclei_Texture_DifferenceVariance_ER_10_0', 'Nuclei_Texture_DifferenceVariance_ER_3_0', 'Nuclei_Texture_DifferenceVariance_ER_5_0', 'Nuclei_Texture_DifferenceVariance_Mito_10_0', 'Nuclei_Texture_DifferenceVariance_Mito_3_0', 'Nuclei_Texture_DifferenceVariance_Mito_5_0', 'Nuclei_Texture_DifferenceVariance_RNA_10_0', 'Nuclei_Texture_DifferenceVariance_RNA_3_0', 'Nuclei_Texture_DifferenceVariance_RNA_5_0', 'Nuclei_Texture_Entropy_AGP_10_0', 'Nuclei_Texture_Entropy_AGP_3_0', 'Nuclei_Texture_Entropy_AGP_5_0', 'Nuclei_Texture_Entropy_DNA_10_0', 'Nuclei_Texture_Entropy_DNA_3_0', 'Nuclei_Texture_Entropy_DNA_5_0', 'Nuclei_Texture_Entropy_ER_10_0', 'Nuclei_Texture_Entropy_ER_3_0', 'Nuclei_Texture_Entropy_ER_5_0', 'Nuclei_Texture_Entropy_Mito_10_0', 'Nuclei_Texture_Entropy_Mito_3_0', 'Nuclei_Texture_Entropy_Mito_5_0', 'Nuclei_Texture_Entropy_RNA_10_0', 'Nuclei_Texture_Entropy_RNA_3_0', 'Nuclei_Texture_Entropy_RNA_5_0', 'Nuclei_Texture_Gabor_AGP_10', 'Nuclei_Texture_Gabor_AGP_3', 'Nuclei_Texture_Gabor_AGP_5', 'Nuclei_Texture_Gabor_DNA_10', 'Nuclei_Texture_Gabor_DNA_3', 'Nuclei_Texture_Gabor_DNA_5', 'Nuclei_Texture_Gabor_ER_10', 'Nuclei_Texture_Gabor_ER_3', 'Nuclei_Texture_Gabor_ER_5', 'Nuclei_Texture_Gabor_Mito_10', 'Nuclei_Texture_Gabor_Mito_3', 'Nuclei_Texture_Gabor_Mito_5', 'Nuclei_Texture_Gabor_RNA_10', 'Nuclei_Texture_Gabor_RNA_3', 'Nuclei_Texture_Gabor_RNA_5', 'Nuclei_Texture_InfoMeas1_AGP_10_0', 'Nuclei_Texture_InfoMeas1_AGP_3_0', 'Nuclei_Texture_InfoMeas1_AGP_5_0', 'Nuclei_Texture_InfoMeas1_DNA_10_0', 'Nuclei_Texture_InfoMeas1_DNA_3_0', 'Nuclei_Texture_InfoMeas1_DNA_5_0', 'Nuclei_Texture_InfoMeas1_ER_10_0', 'Nuclei_Texture_InfoMeas1_ER_3_0', 'Nuclei_Texture_InfoMeas1_ER_5_0', 'Nuclei_Texture_InfoMeas1_Mito_10_0', 'Nuclei_Texture_InfoMeas1_Mito_3_0', 'Nuclei_Texture_InfoMeas1_Mito_5_0', 'Nuclei_Texture_InfoMeas1_RNA_10_0', 'Nuclei_Texture_InfoMeas1_RNA_3_0', 'Nuclei_Texture_InfoMeas1_RNA_5_0', 'Nuclei_Texture_InfoMeas2_AGP_10_0', 'Nuclei_Texture_InfoMeas2_AGP_3_0', 'Nuclei_Texture_InfoMeas2_AGP_5_0', 'Nuclei_Texture_InfoMeas2_DNA_10_0', 'Nuclei_Texture_InfoMeas2_DNA_3_0', 'Nuclei_Texture_InfoMeas2_DNA_5_0', 'Nuclei_Texture_InfoMeas2_ER_10_0', 'Nuclei_Texture_InfoMeas2_ER_3_0', 'Nuclei_Texture_InfoMeas2_ER_5_0', 'Nuclei_Texture_InfoMeas2_Mito_10_0', 'Nuclei_Texture_InfoMeas2_Mito_3_0', 'Nuclei_Texture_InfoMeas2_Mito_5_0', 'Nuclei_Texture_InfoMeas2_RNA_10_0', 'Nuclei_Texture_InfoMeas2_RNA_3_0', 'Nuclei_Texture_InfoMeas2_RNA_5_0', 'Nuclei_Texture_InverseDifferenceMoment_AGP_10_0', 'Nuclei_Texture_InverseDifferenceMoment_AGP_3_0', 'Nuclei_Texture_InverseDifferenceMoment_AGP_5_0', 'Nuclei_Texture_InverseDifferenceMoment_DNA_10_0', 'Nuclei_Texture_InverseDifferenceMoment_DNA_3_0', 'Nuclei_Texture_InverseDifferenceMoment_DNA_5_0', 'Nuclei_Texture_InverseDifferenceMoment_ER_10_0', 'Nuclei_Texture_InverseDifferenceMoment_ER_3_0', 'Nuclei_Texture_InverseDifferenceMoment_ER_5_0', 'Nuclei_Texture_InverseDifferenceMoment_Mito_10_0', 'Nuclei_Texture_InverseDifferenceMoment_Mito_3_0', 'Nuclei_Texture_InverseDifferenceMoment_Mito_5_0', 'Nuclei_Texture_InverseDifferenceMoment_RNA_10_0', 'Nuclei_Texture_InverseDifferenceMoment_RNA_3_0', 'Nuclei_Texture_InverseDifferenceMoment_RNA_5_0', 'Nuclei_Texture_SumAverage_AGP_10_0', 'Nuclei_Texture_SumAverage_AGP_3_0', 'Nuclei_Texture_SumAverage_AGP_5_0', 'Nuclei_Texture_SumAverage_DNA_10_0', 'Nuclei_Texture_SumAverage_DNA_3_0', 'Nuclei_Texture_SumAverage_DNA_5_0', 'Nuclei_Texture_SumAverage_ER_10_0', 'Nuclei_Texture_SumAverage_ER_3_0', 'Nuclei_Texture_SumAverage_ER_5_0', 'Nuclei_Texture_SumAverage_Mito_10_0', 'Nuclei_Texture_SumAverage_Mito_3_0', 'Nuclei_Texture_SumAverage_Mito_5_0', 'Nuclei_Texture_SumAverage_RNA_10_0', 'Nuclei_Texture_SumAverage_RNA_3_0', 'Nuclei_Texture_SumAverage_RNA_5_0', 'Nuclei_Texture_SumEntropy_AGP_10_0', 'Nuclei_Texture_SumEntropy_AGP_3_0', 'Nuclei_Texture_SumEntropy_AGP_5_0', 'Nuclei_Texture_SumEntropy_DNA_10_0', 'Nuclei_Texture_SumEntropy_DNA_3_0', 'Nuclei_Texture_SumEntropy_DNA_5_0', 'Nuclei_Texture_SumEntropy_ER_10_0', 'Nuclei_Texture_SumEntropy_ER_3_0', 'Nuclei_Texture_SumEntropy_ER_5_0', 'Nuclei_Texture_SumEntropy_Mito_10_0', 'Nuclei_Texture_SumEntropy_Mito_3_0', 'Nuclei_Texture_SumEntropy_Mito_5_0', 'Nuclei_Texture_SumEntropy_RNA_10_0', 'Nuclei_Texture_SumEntropy_RNA_3_0', 'Nuclei_Texture_SumEntropy_RNA_5_0', 'Nuclei_Texture_SumVariance_AGP_10_0', 'Nuclei_Texture_SumVariance_AGP_3_0', 'Nuclei_Texture_SumVariance_AGP_5_0', 'Nuclei_Texture_SumVariance_DNA_10_0', 'Nuclei_Texture_SumVariance_DNA_3_0', 'Nuclei_Texture_SumVariance_DNA_5_0', 'Nuclei_Texture_SumVariance_ER_10_0', 'Nuclei_Texture_SumVariance_ER_3_0', 'Nuclei_Texture_SumVariance_ER_5_0', 'Nuclei_Texture_SumVariance_Mito_10_0', 'Nuclei_Texture_SumVariance_Mito_3_0', 'Nuclei_Texture_SumVariance_Mito_5_0', 'Nuclei_Texture_SumVariance_RNA_10_0', 'Nuclei_Texture_SumVariance_RNA_3_0', 'Nuclei_Texture_SumVariance_RNA_5_0', 'Nuclei_Texture_Variance_AGP_10_0', 'Nuclei_Texture_Variance_AGP_3_0', 'Nuclei_Texture_Variance_AGP_5_0', 'Nuclei_Texture_Variance_DNA_10_0', 'Nuclei_Texture_Variance_DNA_3_0', 'Nuclei_Texture_Variance_DNA_5_0', 'Nuclei_Texture_Variance_ER_10_0', 'Nuclei_Texture_Variance_ER_3_0', 'Nuclei_Texture_Variance_ER_5_0', 'Nuclei_Texture_Variance_Mito_10_0', 'Nuclei_Texture_Variance_Mito_3_0', 'Nuclei_Texture_Variance_Mito_5_0', 'Nuclei_Texture_Variance_RNA_10_0', 'Nuclei_Texture_Variance_RNA_3_0', 'Nuclei_Texture_Variance_RNA_5_0'

The extracted features are stored in 406 matrices (406 plates), and I have an index to map metadata (plate_id, well_id, site_id, compound_id) to each row across 406 matrices.

Assay Selections

Three criteria to select our assays:

• Suitable compound size to run advanced CNN models
• Diverse compounds
• Good prior model performance

My plan is to select the top N diverse assays between the first quartile and the second quartile.

Currently, I am trying to use the Tanimoto similarity as discussed in the "Repurposing High-Throughput" paper to measure a set's compound diversity. According to this blog post, I think the average of all pairwise Tanimoto similarities within one group is an okay statistic to measure the assay compound diversity. However, computing all pairwise similarities is very slow (O(N^2)).

We can discuss alternative ways to measure group compound diversity.

Clustered Cross-validation

"Repurposing High-Throughput" discussed it is unfair to randomly assign cross-validation folds, because similar compounds might be assigned into both train and test sets.

Instead, we clustered the compounds based on chemical similarity and randomly assigned the clusters into folds. Here, a Tanimoto similarity cutoff on ECFP6 features was used to ensure close analogs ended up in the same test or validation fold.

For our single-task cross-validation scheme, I guess we can:

1. Cluster the compounds in this assay
2. Assign each cluster as a cross-validation fold

However, it can introduce different fold size within each assay, and different total fold size across all assays. Is this a good way to set cross-validation scheme?

[agitter]

We should ask Moayad and Ben in Slack for their opinions on these issues because they are working on very similar things and have good expertise. Moayad has been working on fast compound clustering and quantifying chemical diversity. Ben is using DeepChem, which has dataset splitting code that is aware of chemical structures.

[agitter]

To help use decide which quadrants to use in terms of assay size, we could also pick 1-3 assays of different sizes. Perhaps ~500, 2500, and 10000 compounds. Then we could test how long it takes to train a CNN from scratch for those assays to guide on when we actually select assays and prioritize all three criteria above (number of compounds, chemical diversity, baseline performance).

[xiaohk]

2215bfcefb919e80116d275579e0a8344389b6b5 adds results of using Moayad's multi-threading code to compute the average of pairwise dissimilarities across all 212 assays.

• The diversity is quite consistent across all the assays
• I have noticed some difference between Moayad's dissimilarity statistics with the (1 - similarity) where similarity is computed by the function rdkit.DataStructs.FingerprintSimilarity()
• I might write a less optimized code with simple producer-consumer multithreading to compute diversity using rdkit.DataStructs.FingerprintSimilarity()

[agitter]

The difference in similarity is interesting and something we can discuss with Moayad

Because diversity does not vary much, we can ignore it for selecting an assay. The next step will be to choose an assay or two and test out training a CNN.

[xiaohk]

I have discussed the difference with Moayad, and we have solved it.

• The dissimilarity computed using his code and rdkit tanimoto similarity should sum up to 1.
• Two values do not sum up to 1 if we use FingerprintMol() to generate fingerprints (the first figure below).
• Two values do sum up to 1 if we use GetMorganFingerprintAsBitVect(mol, radius=2, nBits=2048) instead (the second figure below).
• The diversity plot in https://github.com/gitter-lab/pharmaco-image/issues/7#issuecomment-479535306 is correct.

[xiaohk]

87974f17a0cf8b405b94dc739b11873ed4257359 adds notebook and script to end-to-end train a CNN for binary classification.

Workflow

1. Select one assay, find all (plate_id, well_id) pairs across 406 plates that use the tested compounds (activity is positive or negative).
2. Download selected plates, pull out and merge channels for all image instances for each selected well.
   • Use multiprocessing to work on n plates in parallel.
   • For each channel with size (520, 696), use padding to make it as a square image with size (696, 696).
   • Merge 5 channels together, so we can get a tensor for each image instance with size (5, 696, 696).
   • Store each tensor as a numpy npz file, encode compound activity in the filename.
   • Compress all extracted tensors and store them in Gluster.
3. Use PyTorch to train a simple LeNet to do single-test binary classification for one given assay.
   • Transfer corresponding image tensors for this assay from Gluster when submitting the job.
   • Randomly split images into training, validation (for early stopping), and test sets with proportion 6:2:2.
   • Use DataLoader to load image tensors in parallel in the runtime.
   • Script has flexible learning rate, batch size, and early stopping patience.
   • Save loss function values for training set, validation set during training. Save test results and trained model.

Result

• It successfully uses GPU on gitter0000 nodes to train on assay 57 (having the smallest number of compounds). It took 2 hours for 500 epochs (not trigger early stopping). The test accuracy is 0.52%. It uses 1027 images in total.

Problems

Assay ID	Compound Size	Number of Plates
57	23	67
192	100	93
187	1528	180
162	4641	184
73	10013	268
23	24626	398

• The task to reorganize and merge images from different plates is one of our bottlenecks. My multiprocessing python code can generate different processes, but each process takes longer time to extract from one plate.
  • Assay 192 took 13 hours to extract image tensors on 8 workers. Assay 57 takes 9 hours to extract tensors on 4 workers.
  • 1 worker uses 270 seconds to extract images from one plate. Each of 4 workers uses 700-1600 seconds.
  • Networking is slow. It takes a long time to transfer large zip files from Gluster to the nodes
• The number of plates adds up very quickly for assay with more compounds
• Do we extract tensors first, or we extract them in the CNN training job? Tensor storage problem.

Solutions

• Divide image extraction jobs (plates) by ourselves, instead of using Python multiprocessing. Run each job on a single core cluster. I have 2TB home directory space now.

[agitter]

CHTC gave more storage space to help with this and also said to transfer data from the home directory when we submit to Cooley.

In the future, we can also do the image extraction for DMSO images. This will create a supplemental set of negative examples that could be added to any of the assay-specific training sets.

The assay 57 test accuracy was due to all negative predictions. We can try class weighting and other techniques to avoid this. We can also do more sophisticated cross validation later such as stratified splitting and/or chemical structure based splits.

We can keep the LeNet architecture to continue this testing. One question is how it will scale as we have more images.

We should also keep in mind that this does not do anything to account for batch effects.

[xiaohk]

f4fc6b3e038c92908bd9b42d48490f49ed0e4319 uses DeepChem's ScaffoldSplitter to split used compounds into train/vali/test sets for one assay with given set percentages (I am using 6:2:2).

1a359e00403e588d10d7d1f14e99691a1428e331 adds class weighting for LeNet training.

Current workflow:

1. Choose one array
2. Extract image tensors 5 x 696 x 696 with compounds that are used in the selected array
   • Submit one CPU job for each plate (e.g. 179 jobs for assay 100)
   • Receive one tarball file containing all used images for each plate
3. Compress all tarball files as a zip file, submit one GPU job to train a LeNet CNN
   1. Unzip the zip file, extract all image tensors from tarball files into one directory
   2. Split image tensors into train/vali/test sets (e.g. 4032/1350/1290 images for assay 100)
   3. Train a LeNet CNN

I have applied this workflow on four new assays (64, 100, 146, 161). They all have about 100 compounds. It took about 2 hours to extract image tensors for one assay, and 2~3 hours to train a LeNet CNN.

For all four assays, the training loss and validation loss change but the training accuracy remains the same throughout all epochs. The trained CNN predicts all test images as negative. The training set positive/negative ratio is 77.53%/22.47% for assay 100.

I am using learning rate = 0.0001, batch size = 32. I will try different hyper-parameters to see if that helps training.

[agitter]

Your workflow seems rigorous and uses best practices from computational chemistry and machine learning. I don't see any obvious ways to improve it. Using the scaffold splitting is the correct approach, but it does make the predictive task harder.

Is it possible that the LeNet CNN is too simple to learn the complex transformation between input images and activity? Would it be worthwhile to add more layers and see if that has any effect?

I am also curious whether we can convince ourselves whether or not there is really any signal in this dataset. One approach we have taken in other projects is to randomly shuffle all the class labels in the training data and see if it affects our ability to fit the data. What would the training loss converge to for the real labels versus randomly shuffled labels? Would that be informative here?

[xiaohk]

Yes, it is possible that the LeNet CNN is too simple to transform image signals to compound activity. This paper used a modified VGG network to train on cropped U2OS single cell images. I might try it after training LeNet on more assays.

Finding evidence to show there is no/few signal in the dataset is hard. If the training loss doesn't change after shuffling the labels, then the model fails to learn anything, but It doesn't necessarily tells that there is no signal in the dataset. The method sounds interesting though. I will try it on some assays with more compounds.

[xiaohk]

The workflow I described in https://github.com/gitter-lab/pharmaco-image/issues/7#issuecomment-511882160 didn't work for assays with more compounds. The extracted image tensors can be ~500GB, so my jobs were killed by the CHTC admin.

After talking to Christina, the correct way is to send those image tensors back to Gluster, and then copy them from the servers at runtime. tekin and gzkdon't support Gluster, and the disk size of gitter nodes is not enough for some assays. We have to think some other ways if we really want to push the assay size.

Assay	assay Size	Combined Size of Extracted Tensors
155	733	69GB
21	658	62GB
13	1261	110GB
185	1973	100GB
66	9147	493GB

For assay 155, 21, 13, 185, LeNet gives all positive predictions. I oversampled negative instances during training.

abb68ed73deab47fc0055195b985c8a65c450f7b adds implementation of the network used in this paper. I will update results when the jobs finish running.

[agitter]

We can spend a little more time probing the training data to understand why the training accuracy is also around 50%. First, we can pick a single positive compound and single negative compound. We train on their respective images without early stopping. The LeNet model should be able to overfit this training data and give perfect training set accuracy. If so, we can gradually increase the size of the training data to understand when it becomes impossible to overfit.

The hypothesis is that there isn't real signal in the training data, the positive and negative images look too similar.