search

WayScience / phenotypic_profiling

Machine learning for predicting 15 single-cell phenotypes from cell morphology profiles
Creative Commons Attribution 4.0 International
1 stars 3 forks source link

Minor tweaks to Sup Figure 8 #59

Closed gwaybio closed 6 months ago

gwaybio commented 7 months ago

I also needed to adjust the data processing to account for updated input data types.

Supplementary Figure 8

subset_featurespace_multiclass_model_eval

Supplementary Figure 8. Evaluating multiclass predictions of single-cell phenotypes using AreaShape and Zernike CellProfiler feature subsets.

(A) Confusion matrices comparing models trained on real data vs. shuffled data with CellProfiler AreaShape and Zernike feature data subsets only. The number in each box represents the total count and the color represents the ratio of count over ground truth label. All data show test set performance. (B) Precision recall curves for all 15 phenotypes. The shuffled baseline models (dashed line) performed very poorly for all phenotypic classes. (C) F1 scores for test set predictions for 15 phenotypes and overall performance across all five feature sets.

review-notebook-app[bot] commented 7 months ago

Check out this pull request on  ReviewNB

See visual diffs & provide feedback on Jupyter Notebooks.

Powered by ReviewNB

gwaybio commented 6 months ago

Updated figure

subset_featurespace_multiclass_model_eval