Skip to content

Releases: mskcc/mimsi

v0.4.5

22 Aug 03:13
a1bcb4c
Compare
Choose a tag to compare
  • Output MSI Indeterminate as a possible MSI Class

v0.4.4

27 Aug 20:31
99a4936
Compare
Choose a tag to compare
  • Add visualization tool for microsatellite instances
  • Remove requirements on dateutil, torchvision
  • Readme and documentation updates

v0.4.3

14 Apr 15:46
a5992ab
Compare
Choose a tag to compare
  • Removes dependency on nose for install to alleviate bioconda build issues. Note that this means if you need to run the test suite you must install nose separately
  • Pins future dependency on version of python

v0.4.2

12 Apr 13:32
551cf0c
Compare
Choose a tag to compare
  • Added conda installation support

v0.4.1

31 Mar 20:54
948ce44
Compare
Choose a tag to compare
  • changes to ensure support for latest python versions
  • added --version command line arg
  • updated requirements.txt

v0.4.0

08 Feb 20:42
334cdc1
Compare
Choose a tag to compare
  • Added MIL attention pooling as an option for increasing the interpretability of the model
  • Updated models after further training
  • Additional tests added
  • Updated pytorch
  • Documentation clean up

v0.3.3

16 Nov 22:47
ef3b4c1
Compare
Choose a tag to compare
  • Force coverage param in Model instantiation to integer rather then string

v0.3.2

16 Nov 21:58
c9d8dd8
Compare
Choose a tag to compare
  • Add latest version of the lower coverage (100x combined tumor/normal) model
  • fix to address coverage parameter mistakenly interpreted as string

v0.3.1

12 Oct 14:24
7658586
Compare
Choose a tag to compare
  • Hotfix to address some missing arguments for the last validation testing call in our custom training script

v0.3.0

20 Jul 20:27
bad3910
Compare
Choose a tag to compare

Merging in the latest MiMSI update, includes:

  • fixes and speed improvements to the bam2tensor model
  • removal of the giant microsatellites.list file in favor of a smaller one limited to only IMPACT specific regions (significantly speeds up data generation from bam files)
  • training/testing improvements and bug fixes
  • updated pre-trained model