Bugfix:
- errors parsing rare MHCII alleles for netMHC tools
Bugfix:
- update NetMHC tool data directory paths
- fix permissions on configuration script (was non-executable)
- minor documentation improvements
Bugfix:
- remove netMHC binary check from foreignness_score function that does not require netMHC binaries
Added functionality:
- add Dockerfile used to build publically available Docker image (andrewrech/antigen.garnish)
- relax parsing of SnpEff annotations to reduce errors across VCFs and/or SnpEff versions
- simplify internal functions that configure netMHC tools
- use Docker to run tests
Breaking changes:
- new filtering criteria based on TESLA: https://www.parkerici.org/research-project/tumor-neoantigen-selection-alliance-tesla/, see ?garnish_antigens
- update Ensembl transcript CDS database to release 101 (any prior versioned transcript release is acceptable)
- remove support for mitochondrial CDS, xlsx input, JAFFA input, PureCN input to focus on peptide/transcript tables and VCFs
- improve SnpEff annotation parsing
Additional updates:
- allow setting thread use via environment variable AG_THREADS
- include percentile rank filtering at default netMHC thresholds
- simplify installation instructions
- remove 7 dependencies
- additional integration tests
- do not export internal functions
- update to include configuration for netMHC dependencies
- support for paired wild-type and mutant protein level input
- added dissimilarity_score
- updated readme citation, link to manuscript repository
- externalized smith-waterman alignment function
- appropriate test updates
- remove deprecated garnish_fitness function
- update to LICENSE
- prepare for the release of the antigen.garnish manuscript
- multiple metrics of antigen quality are now computed
- dissimilarity_score dummy function present
- pulled most unnecessary mclapply loops to prevent parallelization failures
- functionalized foreignness_score code into external function
- garnish_plot, garnish_summary now incorporate antigen quality metrics
- garnish_antigens now returns ranked antigens by quality metrics
- numerous edge case, efficiency updates and bug fixes
- clonality filter added to garnish_affinity
- RNA expression filter added to garnish_affinity
- fitness model implemented in R vs. Python
- garnish_antigens function added to rank neoantigens
- garnish_score function added to return sample level immune fitness summary
- updated summary and plot functions with better output
- added wiki with installation instructions
- numerous edge case bug fixes
- remove biomaRt dependencies
- add ncbi-blast functionality to determine neoantigen near matches
- add ncbi-blast functionality to determine known IEDB matches
- implement antigen fitness model of Lukza et al. Nature 2017
- add summary plots
- improve test coverage
- inverse match against global normal proteins
- improve testing, test formatting
- add prediction deduplication to garnish_affinity
- add gene fusions as a source of neoantigens using garnish_jaffa
- stable API
- initial version