realbigws / deepbound Goto Github PK

Hello, I am trying to run DeepBound but I have been unable to make it work so far. I successfully installed the software and the dependencies and I am able to run the example as in the README file, however when I run the ./oneline_command.sh with my own bam files I get the following error messages:

#total 0 start boundaries, 0 end boundaries
premature_bound/FP14-15h_sorted_sample_cat not found!
model_file parameters/bou_model_MaxAUC_3lab, window_str 75, node_str 100, state_num 3, feat_num 20, feat_range_str , label_weight_str  
input_file premature_bound/FP14-15h_sorted.bou_feat not found!
premature_bound/FP14-15h_sorted_sample_list not found!
summary end boundaries = 0 / 0 / 0 (corrrect / prediction / label), sensitivity = -nan, precision = -nan
summary start boundaries = 0 / 0 / 0 (corrrect / prediction / label), sensitivity = -nan, precision = -nan

I've tried running the three steps separately, and the problem appears to be in the gsamples step. The output file *.ReadCount appears to be empty when I use my own bam files, but not when I use the example.

My bam files are sorted by genomic position, and I cannot understand why the test.bam works but not mine. I am running the software within a Singularity container, with all dependencies installed in the same versions as indicated in the repository (but that shouldn't be a problem since the example works).

Any hint?
Thanks in advance!