Releases: mikessh/migmap
Releases · mikessh/migmap
Java 9 compatibility
Fixes
Post-analysis patch
Improvements
- Remove the
--write-binary
option. Now all post-analysis and processing is performed on MIGMAP clonotype tables- This also allows merging contigs and correcting errors prior to post-analysis
- More efficient implementation of clonotype tree construction under the
Analyze
routine.
Bug fixes and post analysis routines
Improvements & additions
- Modified API to support clonotype table post-analysis routines
- Basic statistical analysis of somatic hypermutations (
Analyze
module, see readme) - Clonotype tree construction (
Analyze
module, see readme)
Fixes
- Fixed exception with P-segment alignment
- Optimized mutation annotation
Known bugs
--by-read
analysis is not working, use1.0.1
version or later
Amino acids for hypermutations and helper routines
Improvements
- Added amino acid representation for hypermutations: one-to-one correspondence for nucleotide representation, amino acid sequences are provided on cumulative level, i.e. a single effect is reported for sets of adjacent hypermutations.
Helper routines
- Correct routine - applies a frequency + parsimony-based algorithm that traverses variant tree and appends erroneous variants to their parent clonotypes.
- An option to use custom database and a routine to annotate custom reference file with so-called V/J reference points.
P-segments and optimization
- Added P-segment reporting
- Static compilation to speed up critical parts of code
- Refactoring for some source-code bottlenecks
Some fixes
- Renamed the tool to MiGMAP
- Finally fixed (semingly) all issues with ambiguous IgBlast mappings
- Fixes for
--details
option - Add some missing columns to by-read and clonotype table output
Detailed region information
- Added an option (
--details
) to export nucleotide and amino acid sequences of CDR and FR regions, as well as receptor mRNA contig. - Fixed some remaining rare bugs related to ambiguous alignment of D segment with IgBlast
Fixes
Solving complex CDR3 cases
Non-canonical CDR3 workaround
- Fix CDR3 end calculation when J segment reference point is at the boundary or J segment conserved residue is absent with downstream J sequence aligned
- Implement CDR3 start calculation in case V-CDR3 region was not identified by IgBlast
- Add "canonical" CDR3 flag which tells if CDR3 starts with C and ends with F/W
- Add corresponding filter to output (
--allow-noncanonical
)
- Add corresponding filter to output (