Where communities thrive


  • Join over 1.5M+ people
  • Join over 100K+ communities
  • Free without limits
  • Create your own community
People
Repo info
Activity
  • Jan 12 17:17

    nellore on master

    adds deprecation note (compare)

  • Nov 30 2020 19:01
    Rimsk commented #85
  • Nov 30 2020 18:58
    Rimsk commented #85
  • Jul 09 2020 23:08

    nellore on master

    adds unit test to alignment_han… (compare)

  • Oct 18 2019 17:59
    muhammed-ali commented #85
  • Oct 09 2019 17:29
    dfermin commented #85
  • Oct 08 2019 13:14
    dfermin commented #85
  • Apr 09 2019 00:16

    BenLangmead on master

    embed parsed_md (compare)

  • Apr 08 2019 21:26

    BenLangmead on master

    make more portable (compare)

  • Apr 08 2019 21:18

    BenLangmead on master

    some attempts to make this scri… (compare)

  • Dec 03 2018 15:25
    gianmaz edited #88
  • Dec 03 2018 14:25
    gianmaz opened #88
  • Mar 15 2018 20:05
    ChristopherWilks opened #87
  • Mar 15 2018 20:01

    ChristopherWilks on master

    switched to use sratoolkit 2.8.… (compare)

  • Mar 04 2018 22:50

    nellore on master

    patches bowtie2-build in travis… (compare)

  • Mar 04 2018 22:11

    nellore on master

    uses bowtie2 2.3.4.1 (compare)

  • Mar 04 2018 21:57

    nellore on master

    specifies samtools version to i… (compare)

  • Mar 04 2018 21:47

    nellore on master

    updates dependencies Merge branch 'master' of https:… (compare)

  • Mar 04 2018 21:39

    nellore on master

    quote rules Merge pull request #86 from Ben… (compare)

  • Mar 04 2018 21:39
    nellore closed #86
abhinav
@nellore
which input line
this will require adding a line to align_reads.py
Julia di Iulio
@juliadiiulio_twitter
Oh no don't worry, it takes several hours before rail chokes, so I think for now I'll just run the set on different instances :)
abhinav
@nellore
would love to figure out why this happens
but i'd also like to know why you're running rail this way
rather than on all samples at once on EMR
Julia di Iulio
@juliadiiulio_twitter
oh ideally I would definitely run all samples at once on EMR
... it just that I am running into those permission issues on AWS, and devops didn't come back to me yet.. and I have to find a way to get the project going
abhinav
@nellore
sorry to hear you're having trouble!
how many samples are you analyzing in total?
Julia di Iulio
@juliadiiulio_twitter
but I agree, that would def be my first choice :)
abhinav
@nellore
and how many reads per sample?
Julia di Iulio
@juliadiiulio_twitter
96 samples
abhinav
@nellore
have you tried taking your credentials
Julia di Iulio
@juliadiiulio_twitter
~50mio reads
abhinav
@nellore
and launching the EMR job from your laptop?
that's the use case we were targeting
Julia di Iulio
@juliadiiulio_twitter
ha I did not !
abhinav
@nellore
it's much easier
Julia di Iulio
@juliadiiulio_twitter
I'll try
abhinav
@nellore
your credentials might work
but you then need to be able to create the default emr roles from your laptop
if they're already set up for your account it may work
Julia di Iulio
@juliadiiulio_twitter
hahah from what I learned so far... nothing is set up for my account :yum: but I'll try!
Ben Strober
@BennyStrobes
Hi @nellore . I'm hoping to use rail to generate exon-exon junction counts for ~200 samples (at ~50 mi reads). As rail takes a fair amount of time to run, I was planning on running multiple batches. And then aggregating the junction data across each of the batches. I just want to check to make sure doing this would give the same quantification as running all the samples in one batch??
abhinav
@nellore
@BennyStrobes it will if you use first-pass junctions
which are exactly the junctions recorded by snaptron
moreover, you can run only the first part of rail
if you specify that the only deliverable you want is jx
that is
use
-d jx
and then the whole thing takes a bit less than half the time it usually does
do you understand what i mean by "first-pass junctions?"
Ben Strober
@BennyStrobes
Hey @nellore, thanks for the quick reply! Thats perfect, as I want to compare my results to the snaptron data. I don't understand what you mean by "first-pass junctions", could you elaborate a bit?
abhinav
@nellore
junctions detected by the aligner in a given sample on a single pass of alignment
if we share junctions across samples, we may find that some junctions undetected in a given sample after a single pass of alignment are detected there on a second pass of alignment
but if you're comparing with snaptron
yes, you can simply use -d jx
Candace Liu
@cliu72
Hello! I am using Rail-RNA to process my RNA-Seq data, and am wondering if there is a way to remove PCR duplicates using UMIs after the alignment step (bowtie2) but before the tsv files are generated? Maybe stop the job after the bam files are created, do filtering, then resume the job? I am trying to use UMI-tools (https://github.com/CGATOxford/UMI-tools) which requires bam files as input to group and deduplicate PCR duplicates.
undiagnosed
@undiagnosed
Hello. I opened issue #85 which has a description of the problem I am having. I tried running again with uncompressed fastq files at it's currently at > 10 hrs stuck on step 4/24 with 0 tasks completed and no cpu utilization. Any idea what's going on? Thanks
Candace Liu
@cliu72
Hi @nellore , I asked a question a week ago about using UMI-tools with Rail. Did you have a chance to look into oit? Thanks!
undiagnosed
@undiagnosed
From the stacktrace, it's stuck on line 565 in align_reads.py, return_code = bowtie_process.wait(). I guess bowtie isn't running for some reason?
abhinav
@nellore
@cliu72 sorry for the late response. what you're asking for would require some nontrivial hacking -- rail doesn't read BAMs to generate TSVs, but rather hadoop intermediates that are in a SAM-like format. it would probably be much easier for you to write scripts that generate rail-like TSVs from groups of BAMs deduplicated by UMI-tools.
Candace Liu
@cliu72
Hi again! I am trying to incorporate Rail into MultiQC (http://multiqc.info/), a cool tool that aggregates summary statistics from different software into a single HTML report. To do so, MultiQC parses through the output for each tool. Is there an output option for Rail that includes summary statistics such as the number of reads unmapped and multimapped? I know I can calculate these using counts.tsv.gz, but MultiQC tries to avoid this kind of calculation as it may have a huge impact on processing time. If it doesn't already exist, such a file may be useful for future releases - just a suggestion. Thanks!
abhinav
@nellore
@cliu72 hey! which additional summary statistics do you want? counts.tsv.gz already provides the numbers of mapped and unmapped reads in each sample
it also tells you how many reads are mapped uniquely
and you can subtract that from the total number of mapped reads to obtain the number of multimappers
Candace Liu
@cliu72
Hi! Thanks for such a quick response. Yeah, I realize I can calculate summary statistics manually, but was just wondering if they were readily available in some sort of final log file (like in STAR). MutliQC tries to avoid calculating these statistics because it slows down the generation of the HTML report.
abhinav
@nellore
hey, yeah