search

taylor-lab / neoantigen-dev

neoantigen prediction from WES/WGS
4 stars 1 forks source link

Bugfix/hgvsc stringparsing #13

Open anoronh4 opened 3 years ago

anoronh4 commented 3 years ago

Addressing error mentioned in #7 Also changing the neoantigen-docker.config file in order to include updated terms used in the newest version of neoantigen.py. Question: i noticed that in the default config file the algorithm versions are changed to 4.0 and 4.1b, should we match the docker config?

gongyixiao commented 3 years ago

I agree with this solution. Let's ask @cband and then proceed

One thing need to be noticed is that Neoantigen is before SomaticFacetsAnnotation, so changing this will result all the maf to be regenerated, although most of them will remain unchanged.

For the question, do you mean this? https://github.com/mskcc/tempo/blob/5c15fc506edbe77479f07999b7035adbc10efe73/containers/neoantigen/Dockerfile#L6-L7

anoronh4 commented 3 years ago

I agree with this solution. Let's ask @cband and then proceed

One thing need to be noticed is that Neoantigen is before SomaticFacetsAnnotation, so changing this will result all the maf to be regenerated, although most of them will remain unchanged.

For the question, do you mean this? https://github.com/mskcc/tempo/blob/5c15fc506edbe77479f07999b7035adbc10efe73/containers/neoantigen/Dockerfile#L6-L7

yes, i do mean the lines you highlighted. i'm comparing them to the config file for non-docker runs edited here. i'm not sure if 4.0a is older or newer than 4.0. i would have to change neoantigen-docker.config as well as our docker file.

gongyixiao commented 3 years ago

yes, i do mean the lines you highlighted. i'm comparing them to the config file for non-docker runs edited here. i'm not sure if 4.0a is older or newer than 4.0. i would have to change neoantigen-docker.config as well as our docker file.

Would you mind to check in the docker container if these binary is actually 4.0a by running: /usr/local/bin/netMHC-4.0/netMHC and /usr/local/bin/netMHCpan-4.0/netMHCpan, just to make sure.

I think it's good to go now. You can merge it into master and make the release from master instead of from bugfix/HGVSc_stringparse