Open rotoke opened 7 months ago
If it helps anyone: the alignments producing errors were longer than the default 65520 bp hard-coded in Genrich.h. Increasing this value and recompiling solved the issue. I'd still be interested whether this limit has been set for a particular reason?
Thanks, Roman
Great to hear that you diagnosed and solved the problem! I don't recall setting that limit for any particular reason, except that it is far larger than a typical SAM record for a short read. I didn't anticipate long reads being used in genome enrichment assays.
Hello,
I am analysing a series of multifasta files with ecc_finder, and one of them produces the following error at the Genrich step:
Error! seq_2125767: poorly formatted SAM/BAM record
(I can reproduce the error by runningGenrich -t tmp.sam -o sample.Peak -v
on the sorted SAM file)I managed to trace down the error to two alignments within the SAM file, which come from the longest (94247 bp) and second longest (77851 bp) fasta read respectively. However, I can't see any obvious problem with these entries:
SAM header:
Head of malformatted alignment:
Tail of malformatted alignment:
Do you have any idea what could have gone wrong here?
Thank you very much, Roman