General:
New Features:
pos_range_max argument tofindNovelAlleles andplotNovel. Withpos_range_max, TIgGER takes into account the position inwhich the V sequence alignment ends based on the aligner (usuallypos_range_max="v_germline_end"). Withpos_range_max=NULL, mutation count uses all nucleotides inthe IMGT V region. This means that when the V is trimmed on the 3’,TIgGER includes in the mutation count nucleotides from the CDR3.Bug Fixes:
Updated the error check inpositionMutations tocheck for both empty GERM_NT positions and absence of IMGT gaps in thegermline. Before, gapped germlines of less than 312 positions (IGHV4-31*09, 311 positions) would have empty GERM_NT positions, and thefunction would stop with an error message ‘Check you are using gappedreference germlines’.
Fixed bug inselectNovel wherekeep_alleles=T would not keep different alleles leading tothe same novel sequence.
Fixed buggenotypeFasta where it wouldn’t findduplicate genes.
General:
Replaced error message with warning in function getMutatedAA, toallow for germlines with N (e.g. IGHV1-45*01)
To identify the closest reference, thegenerateEvidence function will only consider referencegermlines belonging to the same gene segment. This is to avoid an errorwhen the user provides VDJ references, not just V.
Backwards Incompatible Changes:
V_CALL (Change-O) as the default toidentify the field that stored the V gene calls, they now usev_call (AIRR). Scripts that relied on default values(previously,v_call="V_CALL"), will now fail if calls tothe functions are not updated to reflect the correct value for the data.If data are in the Change-O format, the current default valuev_call="v_call" will fail to identify the column with the Vgene calls as the columnv_call doesn’t exist. In thiscase,v_call="V_CALL" needs to be specified in the functioncall.findNovelAlleles are now using lower case(germline_call, notGERMLINE_CALL)General:
AIRRDb.Dependencies:
Bug Fixes:
sortAlleles that was not sorting correctlyTR gene names.positionMutations that was counting. as mutated position.New Features:
GermlineIGHV and moved old annotations toSampleGermlineIGHV.v_call), J call (j_call),sequence alignment (seq), junction (junction)and junction length (junction_length) in all functions thatuse this information.reassignAlleles withnon-existentv_call column.generateEvidence that was reporting aminoacids mutations as NA instead of gaps.Bug Fixes:
reassignAlleles occuring with singlematch genotypes.selectNovel improperly removing all identicalnovel alleles, rather than keeping a single entry.genotypeFasta will now retain IMGT-numbering spacers as. characters instead of converting them to-characters.findNovelAlleles causing overlyaggressive minimum sequence threshold filtering.getPopularMutationCount.New Features:
inferGenotypeBayesian function.generateEvidence to build a completeevidence table from the results offindNovelAlleles,inferGenotype,inferGenotypeBayesian, andreassignAlleles.findNovelAlleles and adjusted the definitions/names of someexisting columns.keep_gene argument ofreassignAlleles to provide options for maintainingreassignments at the gene (previousTRUE behavior), family,or repertoire level.findNovelAlleles.Backwards Incompatible Refactors:
germline_ighv,sample_db,genotype andnovel_dftoGermlineIGHV,SampleDb,SampleGenotype andSampleNovel,respectively.novel_df argument tonovel inselectNovel,inferGenotype, andgenotypeFasta.novel_df_row argument tonovel_row inplotNovel.inferGenotype was alter forclarity.reassignAlleles so thatit returns the input data.frame with theV_CALL_GENOTYPEDcolumn appended or overwritten.cleanSeqs will no longer replace.characters with-.clip_db todata infindNovelAlleles,plotNovel,inferGenotype andreassignAlleles.findNovelAlleles.inferGenotype would break whenperforming check for alleles that could not be distinguished.inferGenotype would break if allsequences submitted were from a single gene andfind_unmutated was set toTRUE.findNovelAlleles() was not runningin parallel, even whennproc > 1.nproc=1 infindNovelAlleles().