February 2012 - galaxy-user - lists.galaxyproject.org

How can I tell what programs are used by a tool?
by Beale, Holly (NIH/NHGRI) [F] 07 Feb '12

07 Feb '12

Hi all -- I'd like to know what programs are used when I run a tool on my data. Ideally I'd also like to know what parameters are passed to command line tools. For example, I uploaded a sam file to the main galaxy server at PSU. I ran Filter Indels. I got output. What programs were used in this manipulation? samtools mpileup? GATK? FASTX toolkit? I hope I'm missing something extremely obvious. Thanks for your help. Holly

2 2

about Mapping Quality
by Li, Jilong (MU-Student) 07 Feb '12

07 Feb '12

Hi all, I used TopHat to map RNA-Seq reads to genomes. In the output (.sam) file, the value of some mapping quality (the 5th column) is 255. What does it mean? And I found these reads which have mapping quality 255 mapped to unique place. Thanks! Victor

2 1

Re: [galaxy-user] Question about uploading custom index for bowtie
by Jennifer Jackson 07 Feb '12

07 Feb '12

Hi William, To be sure that we are talking about the same thing, you are trying to use a custom reference genome with Bowtie? As the target genome? If so, then what you want to load is a fasta file of the chromosomes/scaffolds/contigs that represent that species, for a particular genome build. This may come from a public source or from your own project, if you are assembling a genome. But regardless of source, the final product should be a fasta file, which you load into a history and label as fasta, and that is what Galaxy would accept as a valid custom reference genome input on tool forms. If all of this sounds like what you are wanting to do, then I am not clear where the fastq data fits in. The query in a Bowtie job would be in fastq format, but no indexes are needed for a query. Indexes should never be datasets in a history. Now, if you are doing something different, such setting up databases in a local install, then creating indexes would be part of that process, but the Galaxy UI is not the place to this. Instead, you will want to follow the procedure in this wiki: http://wiki.g2.bx.psu.edu/Admin/NGS%20Local%20Setup (in particular, see "Setting Up the Reference Genomes for NGS Tools") All of that said, there are no specific blocks that would prevent you from using something other than a full reference genome as a target in a Bowtie run. I am fairly certain that almost any properly formatted fasta file could be used as long as the sequence identifiers were unique within the file. I've created and used small "dummy" fasta files as custom reference genomes with Bowtie myself without issues. My apologies if your question is still not addressed. So if this does not clear things up, maybe it would help if you explained a bit more about what the final goal is? We want to make sure you are getting the correct help to resolve any outstanding issues, Best, Jen Galaxy team On 2/7/12 12:34 PM, William Light wrote: > Jennifer, > > I am having some issues figuring out if I can use bowtie to take my > index (I have two that I want to upload) and generate one fasta file. > When I use the bowtie-inspect command, the only thing I seem able to > generate is a list of the sequences used to make the index (which are > SOLiD sequences that I used to make the index) rather than one fasta > file that represents the seqeunce. I have the fastq that I used to > generate the reference uploaded right now, and in poking around, it > seems that there is some issue with making the .ebwt files on a platform > other than the one that galaxy runs on. Am I just missing the right > command line settings for making one fasta file? > > - Will Light > Northwester University > > On Mon, Feb 6, 2012 at 8:14 PM, Jennifer Jackson <jen(a)bx.psu.edu > <mailto:jen@bx.psu.edu>> wrote: > > Hello William, > > To use a custom reference genome, all you need to upload is the > single fasta file for the genome. Galaxy will do the rest and create > indexes as appropriate. > > Hopefully this helps! > > Thanks, > Jen > Galaxy team > > > On 2/6/12 11:51 AM, William Light wrote: > > I recently tried to upload a custom made index for bowtie using > Filezilla as my FTP source, but I got an error message, I think > due to > the autodetect for file type not recognizing this file type. Is > there > something special that I should do to upload my six .ebwt files > for my > reference? > > > > _____________________________________________________________ > The Galaxy User list should be used for the discussion of > Galaxy analysis and other features on the public server > at usegalaxy.org <http://usegalaxy.org>. Please keep all > replies on the list by > using "reply all" in your mail client. For discussion of > local Galaxy instances and the Galaxy source code, please > use the Galaxy Development list: > > http://lists.bx.psu.edu/__listinfo/galaxy-dev > <http://lists.bx.psu.edu/listinfo/galaxy-dev> > > To manage your subscriptions to this and other Galaxy lists, > please use the interface at: > > http://lists.bx.psu.edu/ > > > -- > Jennifer Jackson > http://usegalaxy.org > http://galaxyproject.org/wiki/__Support > <http://galaxyproject.org/wiki/Support> > > -- Jennifer Jackson http://usegalaxy.org http://galaxyproject.org/wiki/Support

1 0

SamTools Filter Pileup "Only report variants?" option is misleading
by Jesse Erdmann 07 Feb '12

07 Feb '12

In using the filter pileup tool we've discovered that the option "Only report variants?" will throw away lines that have indels provided that they do not contain any SNPs. For example the line: chr1 88427041 C 38 ..+1A..+1A.....+1A.,..+1A.....,.+1A,.,......,...,^~.^~,^~,^~, HHHHHFEHHHHHBFHHHHHHBFHHHEHHHAHHHFHG#E will not be included in the filtered results. At the very least we would recommend relabeling the option to "Only report SNP variants?" But it would also be helpful to include an option to filter for both SNP and indel variants. Thanks! -- Jesse Erdmann Bioinformatics Analyst Masonic Cancer Center University of Minnesota jerdmann(a)umn.edu 612-626-3123

2 1

Re: [galaxy-user] Disk Quote
by Zack Liu 06 Feb '12

06 Feb '12

Dear galaxy admins, I am working on a hi-seq project, where each file is ~ 50G. I have 8 of them. After I uploaded my files, I realized that I have reach my quote limit. Literally, I can't do any mapping or actions on these files since I have no disk space for it. I was wondering if there's anyway I can get more disk storage on galaxy. Thanks! Zack Liu

2 1

Question about uploading custom index for bowtie
by William Light 06 Feb '12

06 Feb '12

I recently tried to upload a custom made index for bowtie using Filezilla as my FTP source, but I got an error message, I think due to the autodetect for file type not recognizing this file type. Is there something special that I should do to upload my six .ebwt files for my reference?

2 1

RNA-Seq analysis
by 杨继文 06 Feb '12

06 Feb '12

Since the pictures are too big for the mailling list. I will upload it in seperate emails. Dear all, I am using Galaxy for RNA-Seq analysis. I expect two lists: differentially expressed transcripts and differentially expressed genes. In these two lists, I would like to see the gene name, gene ID and transcript ID. What I did is: I run cufflinks using reference gene sets (GTF file) from Ensembl (see picture "human Chr19 refgene"). I modified the ensembl GTF file according to http://main.g2.bx.psu.edu/u/jeremy/p/transcriptome-analysis-faq#faq4 so that cufflinks can recognize the column for chromosomes. I got "cufflinks assembled transcript" which shows nicely the gene ID, transcript ID (see picture "Cufflinks assembled transcript"), but the gene name was lost in this file. Then I run cuffcompare using the same reference gene sets (GTF file) from Ensembl. In the output file (picture "cuffcompare combined transcript") you can see that gene name appeared, but galaxy assigned new ID to gene and transcript. Then I run cuffdiff . Output file (see picture "Cuffdiff transcript differential expression") only contains gene name. My question is: how can I keep the information from the reference gene sets during the whole analysis process so that I get meaningful information. Or it is that possible that I retrieve "gene ID, transcripte ID" by using the output file "Cuffdiff transcript differential expression" from cuffdiff? I hope you can help me. Many thanks in advance. JIwen

1 0

RNA-Seq analysis
by 杨继文 06 Feb '12

06 Feb '12

Dear all, I am using Galaxy for RNA-Seq analysis. I expect two lists: differentially expressed transcripts and differentially expressed genes. In these two lists, I would like to see the gene name, gene ID and transcript ID. What I did is: I run cufflinks using reference gene sets (GTF file) from Ensembl (see picture "human Chr19 refgene"). I modified the ensembl GTF file according to http://main.g2.bx.psu.edu/u/jeremy/p/transcriptome-analysis-faq#faq4 so that cufflinks can recognize the column for chromosomes. I got "cufflinks assembled transcript" which shows nicely the gene ID, transcript ID (see picture "Cufflinks assembled transcript"), but the gene name was lost in this file. Then I run cuffcompare using the same reference gene sets (GTF file) from Ensembl. In the output file (picture "cuffcompare combined transcript") you can see that gene name appeared, but galaxy assigned new ID to gene and transcript. Then I run cuffdiff . Output file (see picture "Cuffdiff transcript differential expression") only contains gene name. My question is: how can I keep the information from the reference gene sets during the whole analysis process so that I get meaningful information. Or it is that possible that I retrieve "gene ID, transcripte ID" by using the output file "Cuffdiff transcript differential expression" from cuffdiff? I hope you can help me. Many thanks in advance. JIwen

1 0

Tophat
by 杨继文 03 Feb '12

03 Feb '12

Dear all, Today I used Tophat to map the reads from RNA-Seq, and had a look at the results using "visaulization" function. In some regions I can see splice junctions calculated by Tophat, but I don't see any reads mapping to this region. Is this a bit strange? I thought "splice junction" is calculated from the mapped reads. However, I think cufflinks only uses "accepted hits" as input, maybe it is not a big problem that "splice junction' is not accurate. Am I right? Was there something wrong when I mapped my reads with tophat? Hope you can help me figure it out. Looking forward to hearing from you. Jiwen

2 1

Condor in bioinformatics pipeline
by Nathan Panike 03 Feb '12

03 Feb '12

Has anyone had success using the Condor scheduling system with the following software? RSEM ChimeraScan GATK BreakDancer VariationHunter Thanks, Nathan Panike, nwp(a)cs.wisc.edu UW-Madison Center for High Throughput Computing

1 0