Open alephreish opened 9 months ago
agat does not crash but the translation is incorrect:
$ agat_sp_extract_sequences.pl -g genome.gff -f genome.fna -t cds -p -o genome.faa
$ seqkit grep -rp BM212_RS11370 genome.faa
>gene-BM212_RS11370 gene=nbis-gene-2141 seq_id=NZ_FOSJ01000070.1 type=cds
MSTRRPRRTYTEEFKKQIVDLHKAGKSRKEIIEEYDLTGSAFDKWVRQHNQTGSFKEKDN
LTPEQKELKELRKANTQLQMENDILKQAALIFGRKSK*SNVININILYQRCAVPLRSVED
LIIMK**RKKVTRNLNKRLLKSSLKAKTIMARVN*KND*RNVRLLCPVGGLGAL*TSFIW
YPTMINGHTNRNRVASIKRRLKTH*TESSIKSSR*KQSSRT*PMSKLPTSGSMSVLF*TY
LTARLSGIPQGQIRQLTWFCRLSLQLRETYIRSRYSILTGERNSTTTS*MNYWVPSILCA
R*VEKEILTIMP*RNLRINHLSLNLSTVTHSIHSMNCKSN*WTMFIGGILFAHMDPWTTK
RLSIIVKVGNRNSLN
I do not entirely understand what is happening, but in some cases CDSs with ribosomal slippage from NCBI cause
double free or corruption (!prev)
:Notwithstanding the crash, the CDS is translated correctly:
If translation is not requested, there is no crash but gffread produces weird CDS features:
Interestingly, there is a different transposase CDS in this example with ribosome slippage that does not cause the crash:
But gffread still corrupts the gff in this case as well: