SchulzLab / STARE

TF analysis from epigenetic and Hi-C data
MIT License
16 stars 2 forks source link

Possible minor bugs in runTestCases.sh #8

Closed MattQIMR closed 1 week ago

MattQIMR commented 8 months ago

There seems to be minor bugs in your runTestCases.sh.

TestV16: ABC 500kb - Multiple Signal Columns - transfac PSCM file with automatic base content - Binary output Preprocessing region file Transforming transfac-PSCMs to PSEMs Starting TRAP 0s TRAP ABC-scoring region-gene interactions Reading annotation and intersecting peaks 0s ABCpp: intersected genes and peaks 21 mapped to chr21_contacts.txt.gz Fit for pseudocount not possible, proceeding without pseudocounts. 6ms; 21 contact matrix read 2ms; 21 processed and written Compressing files 65ms compression 0s ABCpp in total Generating TF-Gene scores TRAP file read Get the gene-region mapping from the ABC-scoring file(s) Fetch affinity scalings out of the region file. affinities fetched Summarize the region affinities on gene level. 0ms; Scores processed Compressing output 7ms; Compression 0s TF-Gene Scores Congratulations it worked!

TestV17: separate ABC scoring - Multiple Signal Columns - average TSS Reading annotation and intersecting peaks ERROR Can't open the output path, please check carefully: /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V17/Test_V17_ABCpp_Temp_GeneWindow.bed find: ‘/mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V17’: No such file or directory ls: cannot access /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V17/*: No such file or directory ERROR: the number of files does not agree: /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V17 /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Test//Test_Controls/Test_V17

TestV18: separate ABC scoring - Use contact estimate - average TSS Reading annotation and intersecting peaks ERROR Can't open the output path, please check carefully: /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V18/Test_V18_ABCpp_Temp_GeneWindow.bed find: ‘/mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V18’: No such file or directory ls: cannot access /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V18/*: No such file or directory ERROR: the number of files does not agree: /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Output//Test_V18 /mnt/lustre/working/lab_tracyo/mattheBa/Packages/Stare/Test//Test_Controls/Test_V18

TestV19: Windows 3kb - Promoter windows instead of bed-file No bed-file provided, will run on promoter windows Preprocessing region file Writing promoter windows Feature (1:98499-101500) beyond the length of 1 size (96000 bp). Skipping. Transforming transfac-PSCMs to PSEMs Starting TRAP 1s TRAP Generating TF-Gene scores TRAP file read affinities fetched Summarize the region affinities on gene level. 1ms; Scores processed Compressing output 10ms; Compression 0s TF-Gene Scores Congratulations it worked! < For Tests 12, 13, 14, 17 & 18 it tries to create a directory in the -i ./Test/ Directory but these already exists, it then tries saving to the -o ./output/Test_X directory which doesn't exist and then can't load from ./output/Test_X.

I believe in this case it's supposed to be creating ./output/Test_X. I fixed it on my end by replacing mkdir "$test_dir"/Test_V12 mkdir "$test_dir"/Test_V13 mkdir "$test_dir"/Test_V14 mkdir "$test_dir"/Test_V17 mkdir "$test_dir"/Test_V18 with mkdir "$output"/Test_V12 mkdir "$output"/Test_V13 mkdir "$output"/Test_V14 mkdir "$output"/Test_V17 mkdir "$output"/Test_V18

DennisHeck commented 8 months ago

Thanks a lot for pointing that out and even give the solution to it! You're right, the mkdir had the wrong path. I pushed the correct version, and it will be included in the next release. Best wishes, Dennis