search

wangjr03 / PPI-based_prediction_enh_gene_links

Algorithm to predict long-range enhancer-gene links based on protein-protein interactions using random forest models
2 stars 0 forks source link

KeyError: "['Unnamed: 0'] not found in axis" #1

Open hcph opened 1 year ago

hcph commented 1 year ago

Hello, I download the TF ChIP from ENCODE and rename the narrowbed file as EZH2-ENCFF610BYU.bed. This file was putted into the folder named TFchip, while after I run the following command, the error comes: "KeyError: "['Unnamed: 0'] not found in axis"". Could you please help me to fix it? Thank you very much! The command is : python ProTECT.py -c "../data/hc_TF_community.txt" \ -t ../TFchip \ -fn ../data/GM12878_feature_names.csv -fm ../data/GM12878_training_matrix_DNase.txt \ -p ../data/GM12878_pos_sample_tf_overlapping.txt \ -n ../data/GM12878_neg_sample_tf_overlapping.txt \ -l ../data/GM12878_label_list.txt -o ../ -s test

haowang0508 commented 1 year ago

Hi,

Thanks for using our software!

There is a couple of issues here:

  1. ProTECT uses PPI to predict enhancer-gene links. This requires TF ChIP-seq for ALL DESIRED TFs. 1 TF ChIP-seq dataset is too few for prediction.
  2. If you would like to use your own TF ChIP-seq datasets, please go through the 'Generate your own data' section. This requires changes in files used in -fn -p -n.

Thanks

hcph commented 1 year ago

Hi,

Thanks for using our software!

There is a couple of issues here:

  1. ProTECT uses PPI to predict enhancer-gene links. This requires TF ChIP-seq for ALL DESIRED TFs. 1 TF ChIP-seq dataset is too few for prediction.
  2. If you would like to use your own TF ChIP-seq datasets, please go through the 'Generate your own data' section. This requires changes in files used in -fn -p -n.

Thanks

Ok, I will try. Thank you very much!

haowang0508 commented 1 year ago

Also I modified the code a little bit to fix the format issue. Could you please pull the repo again and have a try?

hcph commented 1 year ago

Also I modified the code a little bit to fix the format issue. Could you please pull the repo again and have a try?

That's great! Thank you very much. I will try.