Skip to content

Dedup does not work from std/in #55

Open
@slazicoicr

Description

@slazicoicr

I have a small MiSeq bam file (in.bam) that passes the bamutil validate --in in.bam --so_coord check:

Number of records read = 44491
Number of valid records = 44491

TotalReads	44491.00
MappedReads	44251.00
PairedReads	44491.00
ProperPair	43249.00
DuplicateReads	0.00
QCFailureReads	0.00

MappingRate(%)	99.46
PairedReads(%)	100.00
ProperPair(%)	97.21
DupRate(%)	0.00
QCFailRate(%)	0.00

TotalBases	6645701.00
BasesInMappedReads	6609701.00
Returning: 0 (SUCCESS)

When I run with direct file input bamutil dedup --in in.bam --out del.bam --verbose it works as expected

Writing del.bam
Successfully marked 9 unpaired and 46 paired duplicate reads

When I try piping the SAM, it detects no duplicates samtools view -h in.bam | bamutil dedup --in - --out del.bam --verbose:

Writing del.bam
Successfully marked 0 unpaired and 0 paired duplicate reads

It throws an error when I try to pipe uncompressed bam file samtools view -hub in.bam | bamutil dedup --in -.ubam --out del.bam --verbose:

Sorting the indices of 101 duplicated records
Exiting due to ERROR:
	FAIL_ORDER: Cannot read header since the file pointer is null

My bamutil version:

Set of tools for operating on SAM/BAM files.
Version: 1.0.14; Built: Thu Apr  4 10:14:43 EDT 2019 by slazic

Metadata

Metadata

Assignees

No one assigned

    Labels

    No labels
    No labels

    Type

    No type

    Projects

    No projects

    Milestone

    No milestone

    Relationships

    None yet

    Development

    No branches or pull requests

    Issue actions