Error in seq_gtf(gtf, seqpath, ...) : unused argument (readlen = 70)

[irenerodriguez]

Hi @alyssafrazee

First of all, thank you for writing this package.

I am trying to simulate some reads from GTF format + FASTA sequence files. I adapted the example provided in https://github.com/alyssafrazee/polyester --> simulating reads.
Everything works fine when I don't use the readlen parameter in the simulate_experiment function, and I get my simulations with read length equals 100 (default).
However, when I use this parameter in the simulate_experiment function (for example, readlen=70), I obtain the following error:

Error in seq_gtf(gtf, seqpath, ...) : unused argument (readlen = 70)

It works:
simulate_experiment(gtf='mygenes.gtf', seqpath=seqpath,
reads_per_transcript=readspertx, num_reps=c(2,2,2),
fold_changes=fold_changes, outdir='output_reads',idfield='name')

It doesn't work:
simulate_experiment(gtf='mygenes.gtf', seqpath=seqpath,
reads_per_transcript=readspertx, num_reps=c(2,2,2),
fold_changes=fold_changes, outdir='output_reads',idfield='name',readlen=70)

Thank you for your help.

Best
Irene.

[alyssafrazee]

Hey @irenerodriguez - sorry for the delay here! Looking in to this and plan to push out a fix soon. Thanks for your patience.

[peter-sw-tsai]

Hi @alyssafrazee , This is a great package, I'm very keen to try out those bias model you have included, unfortunately I've run into the same issue using version 1.7.1 installed via bioconductor, it looks like none of those additional arguments is working in simulate_experiment()?

Best,
Peter

[nickschurch]

I've also hit this bug after installing the package from bioconductor even though its almost 5 months from the bugfix. I guess this hasn't made it into the bioconductor release yet?

[alyssafrazee]

Yikes, I'm so sorry about that -- I thought we had pushed to BioC, but it doesn't look like it yet. I'll look into it this week. In the meantime, installing from github directly should work. Thanks for your patience.

[JMF47]

@nickschurch Hey Nick, try installing again from the devel branch using BioC. Please let me know if it has since been resolved. Will also update the release branch if confirmed this issue is resolved

source("http://bioconductor.org/biocLite.R")
biocLite("polyester")

[nickschurch]

Thanks. JMF47: I tried this but it doesn't seem to have solved the problem. The version of polyester I have according to sessionInfo() is 1.6.0 - is that the version in the dev branch?