# IRC channel logs

## 2021-11-22.log

back to list of logs

***jackhill is now known as KM4MBG
***KM4MBG is now known as jackhill
<zimoun>hi!
<zimoun>PurpleSym, rekado: I do not see GATK in Guix source. Do you have it somewhere else?
<PurpleSym>zimoun: There’s a binary version in guix-science-nonfree.
<PurpleSym>(Not the current version though, which is fully open source.)
<zimoun>PurpleSym, thanks.
<rekado>zimoun: GATK is difficult to build because of unpackaged Java build systems
<zimoun>rekado: yeah, that’s what I have tried to explain to the users here. And pointed to PyTorch blog to explain them that many things is not so easy if upstream do the task more complicated, somehow. :-)
<civodul>hey there!
<civodul>would you know how to benchmark an Eigen-using package, such as salmon?
<civodul>(or anything that shows up in "guix graph -M1 -t reverse-package eigen | xdot -f fdp -")
<zimoun>civodul: does what inside tests/ suit your needs?
<civodul>zimoun: could be; looks like basic_alevin_test.sh is a benchmark, right?
<civodul>the only problem is that it has the hard-coded directory name of some input data
<civodul>and i don't know where to get it
<civodul>ideas?
<zimoun>Ahah! Indeed. Me neither.
<zimoun>For testing, maybe you can use any FASTQ files.
<zimoun>But indexes are required and so on. I do not know what the easiest.
<zimoun>civodul: basically, https://combine-lab.github.io/salmon/getting_started/
<zimoun>curl ftp://ftp.ensemblgenomes.org/pub/plants/release-28/fasta/arabidopsis_thaliana/cdna/Arabidopsis_thaliana.TAIR10.28.cdna.all.fa.gz -o athal.fa.gz
<zimoun>for the reference and index it with salmon
<zimoun>then grad public data from ftp://ftp.sra.ebi.ac.uk/vol1/fastq/DRR016/DRR0161${i}/DRR0161${i}_1.fastq.gz;
<zimoun>idem for _2.
<civodul>so i would follow the "Quantifying the samples" section, right?
<civodul>(remember i have no idea what i'm talking about :-))
<zimoun>me neither, eating too much Salmon is bad for the Planet ;-)
<zimoun>I would start at «Analyzing your RNA-seq data with Salmon»
<zimoun>with the salmon you have optimized. And compare against the non-optimized one.
<zimoun>’alevin’ applies only on a type of data (single-cell). And I do not know where to fetch such data.
<civodul>okay
<civodul>i'll see what i can do and ask you more questions if needed :-)
<zimoun>’quant’ is much simpler because it applies to RNA-seq data. Which is now routine.
*rekado updates CRAN packages…
<rekado>we have a problem we need to address at a near point in the future: mathjax
<rekado>there are some packages that I haven’t upgraded in a while because they depend on mathjaxr. One of them is now a major version out of date.
<civodul>d'oh
<civodul>"to address" as in "what do we do with JS code"?
<rekado>yes, but also specifically: we gotta make some progress with mathjax