I've run in to the following issues while trying to get MSMutect to work:
- the path to the data folder in shell_for_all.sh needs to be absolute.
- make a symlink with link name phobos in bin, pointing to the correct phobos binary.
- bowtie2 needs to have its indices specified by the -x flag, so this needs to be added to the awk-oneliner invoking bowtie2 for each repeat.
- Even when adding the -x flag bowtie2 complains about some missing reference files (using the Normal.bam) test data.
- on line 211 in Python_managing_file.py vec[:,fi[0]] would sometimes (also on the Normal.bam test data) raise an IndexError, when wrapping this in a try-catch clause the shell_for_all.sh would run without problems for the Normal.bam and Tumor.bam and ms_indel_calling.sh would also run without problems but the resulting maf_like-files would be all empty.
To setup a fitting environment on the milou cluster;
module load bioinfo-tools BEDTools/2.26.0 R/2.15.2 java/sun_jdk1.8.0_92 bowtie2/2.3.2 gcc/4.8 python/2.7
Dear,
I also have several doubts about these tools, and have no documentation. I would also like to discuss how to successfully run msmutsig. For example, I was able to run MSmutect, but I can not reproduce with successfully MSMutsig for WG data and others, the P-value values obtained weren't significant for samples that I previously know are MSI-High. Thanks to all who can help, I am looking forward to your response too,