selectNovel removes all identical novel alleles

[Jason Vander Heiden]

selectNovel will remove both novel alleles when two identical allele calls are present. For example, IGHV1-2*02_T163C and IGHV1-2*05_T299C are both dropped when present. Changing the keep_alleles argument has no effect.

[ssnn]

I think this, based on description above, should reproduce the problem, but it doesn't

novel_df <- data.frame(NOVEL_IMGT=c("CAGGTGCAGCTGGTGCAGTCTGGGGCT...GAGGTGAAGAAGCCTGGGGCCTCAGTGAAGGTCTCCTGCAAGGCTTCTGGATACACCTTC............ACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGACGGATCAACCCTAAC......AGTGGTGGCACAAACTATGCACAGAAGTTTCAG...GGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCCTACATGGAGCTGAGCAGGCTGAGATCTGACGACACGGCCGTGTATTACTGTGCGAGAGA",
                                    NA,
                                    "CAGGTGCAGCTGGTGCAGTCTGGGGCT...GAGGTGAAGAAGCCTGGGGCCTCAGTGAAGGTCTCCTGCAAGGCTTCTGGATACACCTTC............ACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGACGGATCAACCCTAAC......AGTGGTGGCACAAACTATGCACAGAAGTTTCAG...GGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCCTACATGGAGCTGAGCAGGCTGAGATCTGACGACACGGCCGTGTATTACTGTGCGAGAGA"),
                       POLYMORPHISM_CALL=c("IGHV1-2*02_T163C", NA, "IGHV1-2*05_T299C"),
                       GERMLINE_CALL=c("IGHV1-2*02", "IGHV1-2*02", "IGHV1-2*05"))

Output with keep_alleles=T:

> selectNovel(novel_df, keep_alleles = T)
# A tibble: 2 x 3
  NOVEL_IMGT                            POLYMORPHISM_CA… GERMLINE_CALL
  <fct>                                 <fct>            <fct>        
1 CAGGTGCAGCTGGTGCAGTCTGGGGCT...GAGGTG… IGHV1-2*02_T163C IGHV1-2*02   
2 CAGGTGCAGCTGGTGCAGTCTGGGGCT...GAGGTG… IGHV1-2*05_T299C IGHV1-2*05   

Output with keep_alleles=F

> selectNovel(novel_df, keep_alleles = F)
                                                                                                                                                                                                                                                                                                                        NOVEL_IMGT
1 CAGGTGCAGCTGGTGCAGTCTGGGGCT...GAGGTGAAGAAGCCTGGGGCCTCAGTGAAGGTCTCCTGCAAGGCTTCTGGATACACCTTC............ACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGACGGATCAACCCTAAC......AGTGGTGGCACAAACTATGCACAGAAGTTTCAG...GGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCCTACATGGAGCTGAGCAGGCTGAGATCTGACGACACGGCCGTGTATTACTGTGCGAGAGA
  POLYMORPHISM_CALL GERMLINE_CALL
1  IGHV1-2*02_T163C    IGHV1-2*02
> 

[Jason Vander Heiden]

I can't reproduce this now using the attached novel allele data.frame which previously showed the error.

Seems like we fixed it at some point?

[ssnn]

Yes, seems to work fine with the attached selectNovel_error.rda

> selectNovel(novel_ur01, keep_alleles = F)
  GERMLINE_CALL                NOTE POLYMORPHISM_CALL NT_SUBSTITUTIONS
1    IGHV1-2*02 Novel allele found!  IGHV1-2*02_T163C           163T>C
                                                                                                                                                                                                                                                                                                                        NOVEL_IMGT
1 CAGGTGCAGCTGGTGCAGTCTGGGGCT---GAGGTGAAGAAGCCTGGGGCCTCAGTGAAGGTCTCCTGCAAGGCTTCTGGATACACCTTC------------ACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGACGGATCAACCCTAAC------AGTGGTGGCACAAACTATGCACAGAAGTTTCAG---GGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCCTACATGGAGCTGAGCAGGCTGAGATCTGACGACACGGCCGTGTATTACTGTGCGAGAGA
  NOVEL_IMGT_COUNT NOVEL_IMGT_UNIQUE_J NOVEL_IMGT_UNIQUE_CDR3 PERFECT_MATCH_COUNT PERFECT_MATCH_FREQ
1              154                   6                    146                 184          0.7698745
  GERMLINE_CALL_COUNT GERMLINE_CALL_PERC MUT_MIN MUT_MAX MUT_PASS_COUNT
1                 239                1.8       1      10            216
                                                                                                                                                                                                                                                                                                                     GERMLINE_IMGT
1 CAGGTGCAGCTGGTGCAGTCTGGGGCT---GAGGTGAAGAAGCCTGGGGCCTCAGTGAAGGTCTCCTGCAAGGCTTCTGGATACACCTTC------------ACCGGCTACTATATGCACTGGGTGCGACAGGCCCCTGGACAAGGGCTTGAGTGGATGGGATGGATCAACCCTAAC------AGTGGTGGCACAAACTATGCACAGAAGTTTCAG---GGCAGGGTCACCATGACCAGGGACACGTCCATCAGCACAGCCTACATGGAGCTGAGCAGGCTGAGATCTGACGACACGGCCGTGTATTACTGTGCGAGAGA
  GERMLINE_IMGT_COUNT POS_MIN POS_MAX Y_INTERCEPT Y_INTERCEPT_PASS SNP_PASS UNMUTATED_COUNT
1                   0       1     312       0.125                1      213             184
  UNMUTATED_FREQ UNMUTATED_SNP_J_GENE_LENGTH_COUNT SNP_MIN_SEQS_J_MAX_PASS ALPHA MIN_SEQS J_MAX
1      0.7698745                                63                       1  0.05       50  0.15
  MIN_FRAC
1     0.75
> selectNovel(novel_ur01, keep_alleles = T)
# A tibble: 2 x 30
  GERMLINE_CALL NOTE   POLYMORPHISM_CA… NT_SUBSTITUTIONS NOVEL_IMGT  NOVEL_IMGT_COUNT NOVEL_IMGT_UNIQ…
  <chr>         <chr>  <chr>            <chr>            <chr>                  <int>            <int>
1 IGHV1-2*02    Novel… IGHV1-2*02_T163C 163T>C           CAGGTGCAGC…              154                6
2 IGHV1-2*05    Novel… IGHV1-2*05_T299C 299T>C           CAGGTGCAGC…              154                6
# ... with 23 more variables: NOVEL_IMGT_UNIQUE_CDR3 <int>, PERFECT_MATCH_COUNT <int>,
#   PERFECT_MATCH_FREQ <dbl>, GERMLINE_CALL_COUNT <int>, GERMLINE_CALL_PERC <dbl>, MUT_MIN <int>,
#   MUT_MAX <int>, MUT_PASS_COUNT <int>, GERMLINE_IMGT <chr>, GERMLINE_IMGT_COUNT <int>,
#   POS_MIN <int>, POS_MAX <int>, Y_INTERCEPT <dbl>, Y_INTERCEPT_PASS <int>, SNP_PASS <int>,
#   UNMUTATED_COUNT <int>, UNMUTATED_FREQ <dbl>, UNMUTATED_SNP_J_GENE_LENGTH_COUNT <int>,
#   SNP_MIN_SEQS_J_MAX_PASS <int>, ALPHA <dbl>, MIN_SEQS <dbl>, J_MAX <dbl>, MIN_FRAC <dbl>
> 

[ssnn]

Can't reproduce.