diekhans / intronprospector Goto Github PK

Identify putative introns in RNA-Seq alignments

License: Other

Makefile 9.35% C++ 77.88% C 9.83% Awk 0.18% M4 2.76%

intronprospector's Introduction

intronProspector

Identify putative introns in RNA-Seq alignments in BAM/SAM alignments. This works with both short and long-read RNA sequencing.

Usage is described in the intronProspector manual page and intronProspectorMerge manual page

Installation

See INSTALL.md for instation

intronprospector's People

Stargazers

Watchers

Forkers

maximilianh

intronprospector's Issues

generate an error is not stranded junctions are found

require at least some junctions with strand

rethink max-anchor-indel-size default

--max-anchor-indel-size=36 seems too high

score field exceeds 1000

Got this from bedToBigBed:

Error line 7 of hub/introns.sorted.bed: score (1346) must be between 0 and 1000

Line was:
chr12 6534569 6534809 sj6 1346 +

where does the input come from?

The input is by default read from stdin, but that could be mentioned somewhere in the docs.

In addition, it would be nice if one could read the input from a file instead of stdin. Otherwise, I don't see any way to call this without a shell around it...

use sorted BAM to partition extraction and reduce memory usage

only partition by chromosome is easy, as BAM is not sorted by reverse-end coordinates.

add ability to read/write compressed files based on extension

provide a troff man page

troff man page could be generated by pandoc

have option to generate BEDs with extra info in calls TSV

Provide bigBed AS file and option to create bedBed inputs with extra infromation

missing target cause missing output even with -S

intronProspector -C 1.0 -g tmp/hg38.fa -S --map-to-ucsc --intron-calls=merged/HS_A_Brain.introns.tsv.ku.gi.ucsc.edu.26701 --junction-bed=merged/HS_A_Brain.juncs.tsv.ku.gi.ucsc.edu.26701 --intron-bed=merged/HS_A_Brain.introns.bed.ku.gi.ucsc.edu.26701 ../bams/merged/HS_A_Brain.merged.bam


Warning: genomic sequence not found for chrMT splice junctions not available
Warning: genomic sequence not found for GL000225.1 splice junctions not available
Warning: genomic sequence not found for GL000009.2 splice junctions not available
Warning: genomic sequence not found for GL000194.1 splice junctions not available
Warning: genomic sequence not found for KI270742.1 splice junctions not available
Warning: genomic sequence not found for GL000205.2 splice junctions not available
Warning: genomic sequence not found for GL000195.1 splice junctions not available
Warning: genomic sequence not found for KI270733.1 splice junctions not available
Warning: genomic sequence not found for GL000219.1 splice junctions not available
Warning: genomic sequence not found for GL000216.2 splice junctions not available
Warning: genomic sequence not found for KI270712.1 splice junctions not available
Warning: genomic sequence not found for KI270706.1 splice junctions not available
Warning: genomic sequence not found for KI270744.1 splice junctions not available
Warning: genomic sequence not found for KI270734.1 splice junctions not available
Warning: genomic sequence not found for GL000220.1 splice junctions not available
Warning: genomic sequence not found for GL000221.1 splice junctions not available
Warning: genomic sequence not found for KI270711.1 splice junctions not available
Warning: genomic sequence not found for KI270752.1 splice junctions not available
mv: cannot stat ‘merged/HS_A_Brain.introns.tsv.ku-32.kulocal.14169’: No such file or directory

handling conflicting strands in merge

If a bad alignment causes a conflict strand, the first conflict found is discarded with a warning.

This should really keep the majority, but it requires some restructuring

see test case testStrandConflictMerge

would be good to have an example in either the manpage or the help output or both

Examples rock

Clearly document what is required to get strand in output

XS tag
stranded reads
genome sequence