kallisto

[ewels]

Module request: kallisto

[darogan]

Is there anyway to bump this up on your to do list?

I can provide some example data if that helps?

[ewels]

Yeah absolutely - I just need some example logs and then it's very quick.

[darogan]

I attached two kallisto examples run from my new clusterflow pipeline. Which reminds me, do you want my kallisto.cfmod file too?

HS002-PE-R00059_BD0U5YACXX.RHM066_CGATGT_L002_R1.fastq.gz_fastq_kallisto_clusterFlow.txt
HS002-PE-R00059_BD0U5YACXX.RHM067_CAGATC_L002_R1.fastq.gz_fastq_kallisto_clusterFlow.txt

[ewels]

Ok great, thanks! Does Kallisto write any log files itself or is it only this stderr that we have to go on? It's not the nicest log ever..

Would be great to have a module for Kallisto in Cluster Flow - if you could fork and then open a pull request with it that would be brilliant!

[ewels]

So the kallisto docs say that there should be a file called run_info.json which is "a json file containing information about the run". That sounds much easier to parse...

[darogan]

I attached a run_info.json file and pasted below

{
     "n_targets": 176241,
     "n_bootstraps": 100,
     "n_processed": 34503308,
     "kallisto_version": "0.42.5",
     "index_version": 10,
     "start_time": "Sat May 14 20:47:31 2016",
     "call": "kallisto quant -t 1 --pseudobam -i 
/home/rsh46/scratch/Genomes/Homo_sapiens/GRCh38/Homo_sapiens.GRCh38.cdna.all.idx 
-o 
HS002-PE-R00059_BD0U5YACXX.RHM076_CAGATC_L005_R1_val_1.fq.gz_kallisto_output 
-b 100 HS002-PE-R00059_BD0U5YACXX.RHM076_CAGATC_L005_R1_val_1.fq.gz 
HS002-PE-R00059_BD0U5YACXX.RHM076_CAGATC_L005_R2_val_2.fq.gz"
}

But it doesn't include any pseudo-alignment metrics

[quant] finding pseudoalignments for the reads ... done
[quant] processed 34,503,308 reads, 29,228,276 reads pseudoaligned
[quant] estimated average fragment length: 169.568

and then a simple % alignment rate calculated from the above

[ewels]

Damn, that's a shame. A JSON file with a consistent filename is a lot easier to find and parse than a bit of stderr printed somewhere in the middle of some file.

Ok, I'll write the module up to look for the stderr as you say anyway. At least the stderr repeats the input filenames.

[darogan]

Just tested the new kallisto module - works beautifully for me!

Another request though...

Could you add a kallisto plot for the insert sizes from the line

[quant] estimated average fragment length: 169.568

[ewels]

Apologies, that's already parsed but looks like I forgot to add it to the report. As it's a single value I'll probably add it to the General Statistics table instead if that's ok? It should already be in multiqc_data/multiqc_kallisto.txt as well.

[darogan]

Yes, that will work for me

[ewels]

ok great! Let me know if you run into any other problems or would like any new features.

[ale07alvarez]

I would very much appreciate if you could clarify which file from the kallisto output is parse? There are only three output files from kallisto and when I run multiqc .
I get the following

[INFO   ]         multiqc : This is MultiQC v1.0.dev0
[INFO   ]         multiqc : Template    : default
[INFO   ]         multiqc : Searching '.'
[WARNING]         multiqc : No analysis results found. Cleaning up..
[INFO   ]         multiqc : MultiQC complete

thanks!

[ewels]

Hi @ale07alvarez, I replied on your new issue #440.

Phil