calling on single cell rna.seq from bam file generated from cellranger 10x

Alex Lee

unread,

Apr 10, 2018, 1:12:15 PM4/10/18

to Platypus Users

Hi, I just saw paper Zheng et al using freebayes from bam files generated from 10x cellranger, However I'm wondering is it possible to do the same with platypus. I tried running it with the following parameters

but keep getting these errors. I tried increasing maxReadlenght to 5000 but till the same. Any clues? thanks!

Exception OverflowError: 'value too large to convert to short' in 'htslibWrapper.ReadIterator.get' ignored

2018-04-10 10:10:07,710 - WARNING - Found very long read (14952 bases). Capping max read length at --maxSize (1500).

2018-04-10 10:10:07,711 - WARNING - Longer reads will be used for alignments but not to determine window boundaries

2018-04-10 10:10:07,711 - WARNING - Increase --maxSize if you want longer reads to be used for determining widow boundaries

2018-04-10 10:10:07,711 - WARNING - But keep --maxSize below 5000 otherwise problems will occur downstream and may lead to crashes

Exception OverflowError: 'value too large to convert to short' in 'htslibWrapper.ReadIterator.get' ignored

Sven Klages

unread,

Apr 11, 2018, 1:07:16 PM4/11/18

to Platypus Users

How did you run platypus?

A read with 14952 bases is a bit weird for 10xGenomics data (assuming Illumina fastq data).

You should inspect your BAM file and identify the problem with that extremely long read.

Nils K

unread,

Apr 11, 2018, 2:59:09 PM4/11/18

to Platypus Users

I have had similar problems trying to use Platypus to call variants in normal RNA-seq data. From what I could tell, the problem was that reads which spanned an exon-exon junction were aligned with gaps, which Platypus just considered as part of the read length. So if you have a 10kb intron, Platypus sees a read of length 10kb+x and refuses to proceed.

Sorry to say I never actually resolved this, I just ended up using GATK instead.

Nils

Daniel Cooke

unread,

Apr 12, 2018, 9:35:04 AM4/12/18

to Nils K, Platypus Users

Nils,

You might be interested in the RNA-seq preprocessor for Platypus, Opposum, developed in Oxford (https://wellcomeopenresearch.org/articles/2-6/v2). I’ve not used it myself, but I believe it solves the intron problem you mention.

Best,

Dan

--
You received this message because you are subscribed to the Google Groups "Platypus Users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to platypus-user...@googlegroups.com.
To post to this group, send email to platypu...@googlegroups.com.
Visit this group at https://groups.google.com/group/platypus-users.
To view this discussion on the web, visit https://groups.google.com/d/msgid/platypus-users/5c854de3-81e9-4167-8110-da4c7c9c1aa3%40googlegroups.com.
For more options, visit https://groups.google.com/d/optout.

Reply all

Reply to author

Forward