a problem with uploading

[lenis vasilis]

Hello,

i'm trying to upload a small file to the genome browser, the nets from the alignment of the chicken' s chr22 with the Medium ground finch in order to compare it with the existing one to see if my parameters are right.

The problem is that it gives me the following error:

Error File 'chr22.net' - Unrecognized format line 8 of file: net chr22 4081097 (note: chrom names are case sensitive, e.g.: correct: 'chr1', incorrect: 'Chr1', incorrect: '1')

In my file the "chr22" seems to be correct.

A previous time when I had a similar problem I had put a line like this "track name=sample" at the beginning of the file and it worked, but now it continues to complain.

Any idea what can be wrong?

Thank you in advance,

Vasilis.

[Steve Heitner]

Hello, Vasilis.

Can you please send us a copy of your file so we can look at the contents?  You can send it directly to me if you would prefer it not go to the list.

Please contact us again at gen...@soe.ucsc.edu if you have any further questions.  All messages sent to that address are archived on a publicly-accessible Google Groups forum.  If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

---
Steve Heitner
UCSC Genome Bioinformatics Group

--

[Steve Heitner]

Hello, Vasilis.

The problem is that your file is in Net format which is not a supported format for custom tracks.  Please see http://genome.ucsc.edu/goldenPath/help/customTrack.html#format for a list of supported custom track types.

You will want to convert this file to bed format using the netToBed utility which can be obtained from our utilities download page at http://hgdownload.cse.ucsc.edu/admin/exe/.  From there, you will also want to consider additionally converting the bed file into a bigBed file using the bedToBigBed utility in the same directory.  Note that bigBed files are not directly uploaded to UCSC, but rather hosted in a web-accessible location with the URL specified in the bigDataUrl parameter of the track line.  See the bigBed format page at http://genome.ucsc.edu/goldenPath/help/bigBed.html for additional details and examples.

Please contact us again at gen...@soe.ucsc.edu if you have any further questions. 

All messages sent to that address are archived on a publicly-accessible Google Groups forum.  If your question includes sensitive data, you may send it instead to genom...@soe.ucsc.edu.

---
Steve Heitner
UCSC Genome Bioinformatics Group

 

From: lenis vasilis [mailto:va...@aber.ac.uk]
Sent: Wednesday, March 26, 2014 2:15 PM
To: gen...@soe.ucsc.edu
Subject: [genome] a problem with uploading

 

Hello,

--

[lenis vasilis]

Hello,

I’m trying to follow the steps from the wiki page “Whole Genome alignment HowTo” in order to make a pipeline for whole genome alignment.

(I tried to modify the runLastzChain.sh in order to work on an lsf based cluster and I did it, but the alignments that I’m generating are very different with the alignments of the browser. I believe that something I’m missing in the splitting part).

Anyway, at the moment I’m at the lavToPsl step of the alignment HowTo, I created the psl files and I used the pslSwap to change the target- query order.

The problem is that I cannot find the pslSplitOnTarget tool, in order to re-create new psl-files, splited by the name of the target.

I checked to the following link with all the stand alone tools but I cannot find it.

http://hgdownload.cse.ucsc.edu/admin/exe/linux.x86_64.v287/

Could you saw me where can I be able to find it or a way to replace this step of the process?

I’m trying for a long time to make my alignments, so I’m desperately need your support.

Thank you very much in advance,

Vasilis.

[Hiram Clawson]

Good Morning Vasilis:

The procedure you are following that required the pslSplitOnTarget is a procedure
that is being followed due to errors in the preceding steps.

I would highly recommend you follow the steps in runLastzChain.sh which does not
have the procedural error and works the pipeline all the way to chaining. When you
try to combine the procedure with mistakes in it on the "Whole Genome alighment HowTo"
page with the recommendation to use the runLastzChain.sh script, you are confusing
the process. You do not need to mix up these instructions, the runLastzChain.sh script
has all of the procedure in it. There are very few steps in the script and you can
perform each step manually to watch the results accumulate. The chaining step in
the script does not use pslSplitOnTarget:

mkdir -p chain
echo "#!/bin/csh -fe" > chainJobs.csh
for T in `cat target.list | sed -e "s#${WRKDIR}/##"`
do
echo "zcat psl/${T}.*.psl.gz \\"
echo " | axtChain -psl -verbose=0 ${chainParams} \\"
echo -e "\tstdin ${TARGET} ${QUERY} stdout \\"
echo " | chainAntiRepeat ${TARGET} ${QUERY} stdin chain/${T}.chain"
done >> chainJobs.csh

echo "find ./chain -name \"*.chain\" | chainMergeSort -inputList=stdin | gzip -c >
${TNAME}.${QNAME}.all.chain.gz" >> chainJobs.csh

If you really do want to continue the instructions with the error steps, you can
fetch the pslSplitOnTarget source from:
http://genome-source.cse.ucsc.edu/gitweb/?p=kent.git;a=tree;f=src/hg/pslSplitOnTarget

but I do not believe this will help your procedure.

--Hiram