September 2010 - galaxy-user - lists.galaxyproject.org

Python error when running Bowtie for Illumina
by Weng Khong Lim 01 Feb '11

01 Feb '11

Hi all, I'm new to next-gen sequencing, so please be gentle. I've just received a pair of Illumina FASTQ files from the sequencing facility and intend to map them to the hg19 reference genome. I first used the FASTQ Groomer utility to convert the reads into Sanger reads. However, when running Bowtie for Illumina on the resulting dataset under default settings, I received the following error: An error occurred running this job: *Error aligning sequence. requested number of bytes is more than a Python string can hold* * * Can someone help point out my mistake? My history is accessible at http://main.g2.bx.psu.edu/u/wengkhong_lim/h/chip-seq-pilot-batch Appreciate the help! Weng Khong, LIM Department of Genetics University of Cambridge E-mail: wkl24(a)cam.ac.uk Tel: +447503225832

4 3

ABySS
by Jesse Erdmann 08 Oct '10

08 Oct '10

I've done a little searching but haven't found anyone that has wrapped ABySS for Galaxy. Does anyone know of an existing wrapper or is anyone willing to share one? If nothing else it would be nice if we could find a starting spot. Thanks! -- Jesse Erdmann Bioinformatics Analyst Masonic Cancer Center University of Minnesota jerdmann(a)umn.edu 612-626-3123 jesse(a)jesseerdmann.com Twitter: http://twitter.com/jesseerdmann

3 2

searching for data
by Yury Bukhman 08 Oct '10

08 Oct '10

Hi, I would like to annotate my data sets and enable users to search on these annotations. For example, suppose I have datasets from different microbial species and strains, grown on different media under different conditions. I can attach templates to all of my data that would have fields for species, strain, growth medium and treatment. Suppose now I want to look at all of the data from S.cerevisae, strain ABC, grown on a medium XYZ, under heat stress. How do I do that? I have looked around my Galaxy instance and on the Wiki, but I could not figure out how to search on template fields. Thanks. Yury -- Yury V. Bukhman, Ph.D. Associate Scientist, Bioinformatics Great Lakes Bioenergy Research Center University of Wisconsin - Madison 445 Henry Mall, Rm. 513 Madison, WI 53706, USA Phone: 608-890-2680 Fax: 608-890-2427 Email: ybukhman(a)glbrc.wisc.edu

2 1

conditional cluster/merge
by Agnes Williams 08 Oct '10

08 Oct '10

Hi, I have a dataset of some 1000 individuals and I'd like to merge consecutive intervals within 50kb that are within an individual. I know how to do it per person (first filter an individual into one file, and use the "cluster" function at the "operate on genomic intervals". But to do this in 1000 times is obviously not ideal, unless it can be automated. Either that, or perform cluster/merge with an additional condition (only merge for intervals from the same individual). I'm an experimentalist without programming background, so I rely on using galaxy to perform my analyses. I'm wondering if you have any suggestion for this? many thanks in advance! AW

2 1

Getting data from NCBI
by Peter Young 01 Oct '10

01 Oct '10

The Get Data | Get Microbial Data tool is nice, but is limited to a subset of the older genomes among those that are currently available. It should surely be possible to get any accession directly from NCBI, but I don't see a tool to do that. The best I can do is to download a file to my computer and then upload it again to Galaxy. Am I missing something obvious? On a related point, I don't see GenBank listed as a format for conversion. Thanks Peter -- Prof J P W Young Department of Biology (Area 3) University of York Wentworth Way Heslington York YO10 5DD jpy1(a)york.ac.uk tel: +44 1904 328630(direct), 328500(department) fax: +44 1904 328505(department) http://bioltfws1.york.ac.uk/biostaff/staffdetail.php?id=jpwy http://www.york.ac.uk/depts/biol/staff/jpwy/jpwy.htm Email disclaimer: http://www.york.ac.uk/docs/disclaimer/email.htm

3 2

Using output as parameter in workflow
by Lance Parsons 01 Oct '10

01 Oct '10

I have a user who is trying to use the output of summary statistics as input to the next step in a workflow. e.g. using "median + 2*stddev" as the threshold cutoff in the next step, which is "pileup-to-interval". Is that possible in galaxy workflow? The current workaround is to create two separate workflows and input the parameter manually, but that is less than ideal. Any suggestions would be appreciated. -- Lance Parsons - Software Developer/Data Analyst 134 Carl Icahn Laboratory Lewis-Sigler Institute for Integrative Genomics Princeton University

2 1

options for history description using tool config
by Bossers, Alex 01 Oct '10

01 Oct '10

If you configure your tool using the xml config file it displays by default the tool action on data nr in the history description. This is useful but there are option to get more descriptive info in there. Default for the sequence length example: .. <outputs> <data name="output" format="tabular"/> </outputs> .. Gives: 2: Compute sequence length on data 1 This can be altered to have something like a description having the tool settings: .. <outputs> <data name="output" format="tabular" label="SEQLEN of ${query_type.value_label}" /> </outputs> .. Where you can reuse a specific selected value from the form entered by the user. Is there any documentation on this? The default tool_conf xml wiki has no specific info on this. I also noticed the use of .metadata in datasets to specify columns in that data... The specific I want to have for some of my tools is that the history line shows the tool description and the sequence name chosen. If I do: .. <outputs> <data name="output" format="tabular" label="SEQLEN of ${input1}" /> </outputs> .. I do get the sequence FILE name but with the complete not so informative path. Showing the descriptive fasta header would be even nicer.... Thanks Alex

2 1

Fetch Taxonomic Representation on dataset is blank
by Ameena Nalim 27 Sep '10

27 Sep '10

Hello All I ran a metagenomic analysis on a large dataset, larger than any of the others that I have run successfully. Joining two megablast queries resulted in 370, 163, 941 lines in tabular format. I then ran Fetch taxonomic representation on the data It ran with no error messages however when it was done approximately 3 days later, it was a empty file. Why does this happen? The previous dataset with 16,017,326 lines ran successfully If I redo it, how do I know that three days from now, it will not come back empty again? any suggestions? thanks Ameena

2 1

Re: [galaxy-user] integrating Galaxy with a relational data warehouse?
by Yury Bukhman 27 Sep '10

27 Sep '10

Thanks again for your comments. The points about huge file sizes and their "write once" nature are convincing. Are the indexes you are talking about already implemented in Galaxy? Is this how it supports its database-like join and subset operations? What about summarized downstream data types, such as gene intensities, p values from statistical tests etc? Those would seem to be relatively low-volume and less immutable. Suppose, as a simple example, I have a gene expression experiment with several samples (be that arrays or RNA-Seq runs), assigned to 2 treatments. I want to set up a workflow that would first summarize the data to get an expression value for each gene (or exon, or transcribed region) in each run, and then do t tests to discover those that are differentially expressed between the treatments. I'll need to support a project that would perform similarly designed experiments over and over again, e.g. with different cell lines and/or treatments. Although the raw data may remain as flat files in a Galaxy data library, wouldn't it make sense to store the summarized data and t test p values in a relational database? Thanks. Yury On 09/01/10, James Taylor <james(a)jamestaylor.org> wrote: > Exactly. In addition, most relational database are optimized for data that can change, but the access pattern for our raw data is write once. We can implement more efficient storage formats and indexes outside the database for this purpose. > > On Aug 31, 2010, at 5:17 PM, Hiram Clawson wrote: > > > Good afternoon Yury: > > > > Typical file sizes are currently running in the 10s and 100s of Gb for most work flows these days. > > It isn't practical to try and stuff such large single entities into a database. > > It is much more simple to compute indexes into the file and store the indexes in > > the database. We do this all the time at the UCSC genome browser. > > > > --Hiram > > > > Yury Bukhman wrote: > >> Thank you, James, for your reply. I wonder if you could elaborate ... > > _______________________________________________ > > galaxy-user mailing list > > galaxy-user(a)lists.bx.psu.edu > > http://lists.bx.psu.edu/listinfo/galaxy-user > > -- jt > > James Taylor > Assistant Professor > Department of Biology > Department of Mathematics & Computer Science > Emory University -- Yury V. Bukhman, Ph.D. Associate Scientist, Bioinformatics Great Lakes Bioenergy Research Center University of Wisconsin - Madison 445 Henry Mall, Rm. 513 Madison, WI 53706, USA Phone: 608-890-2680 Fax: 608-890-2427 Email: ybukhman(a)glbrc.wisc.edu

2 1

stuck
by Hsu, Amy (NIH/NIAID) [E] 27 Sep '10

27 Sep '10

I'm sure this is a silly question, but I have been stuck since yesterday. I have some whole genome sequencing data - What was given to me is an enormous XL file with all the single base changes identified in 4 people. I don't want to track down all the changes, yet I've noticed some of them labeled as "novel" are in fact SNPs - particularly if I use build 130. I imported just chr, start, end using same base number for start/end from part of my file (chr 1) and then pulled down all the SNPs from UCSC for Chr. 1. What I would like to do is label the lines in my file that are snps. I have tried intersect, join, subtract, all to no avail. What am I doing wrong? Any help would be appreciated. thanks - Amy Hsu

2 1