No fusion found using tests data about genefuse

[sfchen]

I just tried again with command: ./genefuse -r ~/data/ref/hg19.fa -1 ~/data/fq/genefuse.R1.fq.gz -2 ~/data/fq/genefuse.R2.fq.gz -h test.html -j test.json -f genes/druggable.hg19.csv

and got:

15:1:4 start with 4 threads
15:1:47 mapper indexing done
15:2:38 sequence number before filtering: 1329
15:2:38 removeByComplexity: 0
15:2:38 removeByDistance: 39
15:2:38 removeIndels: 67
15:4:3 matcher indexing done
15:4:3 removeAlignables: 8

Probably you used incorrect reference? I used hg19 downloaded from UCSC. Did you checked the downloaded files using MD5?

[dickyornot]

Hi sfchen,

I have the same problem. I used all the demo files you provided including the reference genome. but I still got nothing in my result.

[MarcHiggins]

Hi sfchen,

I have experienced the same problem as others have reported here. Have you gotten to the bottom of this?

Thanks.

[sfchen]

Can you guys check md5 for the downloaded FASTQ file?

[sfchen]

http://opengene.org/dataset.html

You should download following files:
Paired-end FASTQ files for GeneFuse testing (Illumina platform)
genefuse.R1.fq.gz (size: 62 M, MD5: 171e6dfa0af37fe95c826005bc5fcdf9)
genefuse.R2.fq.gz (size: 66 M, MD5: e756cf01e256186dccaa9e700d85a342)

[MarcHiggins]

Hi sfchen,

Yes those are the same md5 I get when I check on the downloaded FASTQs. The command I run is: ./genefuse -r Homo_sapiens_assembly19.fasta -f druggable.hg19.csv -1 genefuse.R1.fq.gz -2 genefuse.R2.fq.gz -h report.html >result

I have downloaded the .fasta file from ensembl.

Thanks.

[sfchen]

The druggable.hg19.csv is in the genes folder

Have you checked the error message?

[sfchen]

I mean, you should run:

./genefuse -r Homo_sapiens_assembly19.fasta -f genes/druggable.hg19.csv -1 genefuse.R1.fq.gz -2 genefuse.R2.fq.gz -h report.html >result

[MarcHiggins]

I have downloaded via wget the druggable.hg.csv from the genes folder. In the results document there are no reported errors

[sfchen]

Errors are saved to STDERR, not STDOUT. So you cannot find errors in the result file.

Can you just run the command without redirecting to result?

[MarcHiggins]

I do not get any STDERR or STDOUT files regardless of if I redirect to result or not. I am running the binary if this may make a difference. Thank you for your help by the way.

[MarcHiggins]

Apologies I meant I do not get an STDOUT file at all.

[sfchen]

You used >, which redirected STDOUT to the file you specified.

[MarcHiggins]

But even if I exclude > there is no STDERR file - that is what I meant not the lack of STDOUT apologies for confusion.

[sfchen]

You didn't redirect STDERR, so it would be printed on terminal.

You can use following command to also redirect STDERR:

./genefuse -r Homo_sapiens_assembly19.fasta -f druggable.hg19.csv -1 genefuse.R1.fq.gz -2 genefuse.R2.fq.gz -h report.html >result & 2>err.log

[MarcHiggins]

I have ran again and no errors are reported. I notice however the FASTQ files don't have the 15 million lines you mention in a different thread - they have more like 800,000. Maybe this is the issue?

[MarcHiggins]

Hi sfchen, I have again ran genefuse on fastqs which I know to contain translocations. Your software did not call these. This is more just to let you know than a specific request or question.

[sfchen]

Thanks, I will try to reproduce it.

[tkcaccia]

Hi sfchen,

I am in the same situation. I do not find genefusion in the test dataset.

[amacbride]

I found the answer in #31 -- the NCBI version of hg19 had a different chromosome naming convention, so it doesn't work. The version downloadable from UCSC is fine:

wget http://hgdownload.cse.ucsc.edu/goldenPath/hg19/bigZips/hg19.fa.gz
(then unzip)