MapSplice2 Problem

[ninni]

Dear all,
The last two months I have been fighting with MapSplice2.
First, we had troubles building the index, but this we fixed. But now, it throws an error regarding the fastqs.

python MapSplice-v2.2.0/mapsplice.py -c human/ensembl/bowtie_index/fasta -x bowtie_index/built_bowtie_index -1 <(zcat R1.fastq.gz) -2 <(zcat R2.fastq.gz) -p 3 -o output

-----------------------------------------------
[Thu Feb 25 09:20:28 2016] Beginning Mapsplice run (MapSplice v2.2.0)
[Thu Feb 25 09:20:28 2016] Bin directory: MapSplice-v2.2.0/bin/
[Thu Feb 25 09:20:28 2016] Preparing output location
[Thu Feb 25 09:20:29 2016] Checking files or directory: /dev/fd/63
[Thu Feb 25 09:20:29 2016] Checking files or directory: /dev/fd/62
[Thu Feb 25 09:20:29 2016] Checking files or directory: /data/human/ensembl/bowtie_index/fasta/
[Thu Feb 25 09:20:29 2016] Checking Bowtie index files
[Thu Feb 25 09:20:29 2016] Inspecting Bowtie index files
[Thu Feb 25 09:20:31 2016] Checking reference sequence length
[Thu Feb 25 09:20:44 2016] Checking consistency of Bowtie index and reference sequence
[Thu Feb 25 09:20:44 2016] Checking read format
-----[Read Format: FASTQ]
-----[Read Type: Pair End]
Base name of two ends not consistent
The 1th read in /dev/fd/63 and /dev/fd/62
FHIIIGI?FAEBFHGGGGGGHEHEEEEECCCCBCBBBBBBBBBB
@HS12_346:7:1101:1516:1746/2 2:N:0:TCGGCA
[MapSplice Running Failed]
Error: Checking read format failed


I do not understand what the problem is. The files work fine with other fusion gene finding tools.

Thanks for any help!

- Ninni

[GenoMax]

Have you compared the two files to ensure that the reads are in order? Has the data been trimmed and if so with what trimmer?

I also noticed that the fastq header in the snippet above does not follow current illumina fastq header convention: @HS12_346:7:1101:1516:1746/2 2:N:0:TCGGCA It appears to be a mix of two formats.

[ninni]

http://seqanswers.com/forums/archive/index.php/t-63770.html

It was this the problem!
I have TCGA data that is why!
Again, thanks, because there you also helped