Kory, Cufflinks quantifies transcripts in parallel, and the inconsistencies that you're seeing are likely the result of these processes finishing in different order during different runs. To address this issue, you can use the Galaxy sort tool. Ultimately, this is not a Galaxy issue but an issue with the Cufflinks tool. You could drop the Cufflinks authors a note if you feel strongly about this inconsistency: tophat.cufflinks@gmail.com Finally, please use the galaxy-user mailing list for questions about tool usage. The galaxy-dev list is for local installation and tool development questions. Best, J. On Mar 21, 2012, at 10:51 AM, Johnson, Kory (NIH/NINDS) [C] wrote:

Hello,

I performed 6 different runs of Cufflinks using the same Reference Annotation on 6 different samples. I expected the resulting tabular output to be constant given the same Reference Annotation was used each time. What I found is that there was an inconsistency.

Specific example, here is top 5 rows from output for one of the six samples:

tracking_id class_code nearest_ref_id gene_id gene_short_name tss_id locus length coverage FPKM FPKM_conf_lo FPKM_conf_hi FPKM_status NM_001005484 - - NM_001005484 - - chr1:69090-70008 - - 0 0 0 OK NR_026820 - - NR_026820 - - chr1:34610-36081 - - 0 0 0 OK NM_001005221 - - NM_001005221 - - chr1:367658-368597 - - 0 0 0 OK NR_046018 - - NR_046018 - - chr1:11873-14408 - - 0 0 0 OK

Here are top 5 rows from output for another of the six samples:

tracking_id class_code nearest_ref_id gene_id gene_short_name tss_id locus length coverage FPKM FPKM_conf_lo FPKM_conf_hi FPKM_status NM_001005484 - - NM_001005484 - - chr1:69090-70008 - - 0 0 0 OK NR_026820 - - NR_026820 - - chr1:34610-36081 - - 0 0 0 OK NM_001005221 - - NM_001005221 - - chr1:367658-368597 - - 0 0 0 OK NR_046018 - - NR_046018 - - chr1:11873-14408 - - 0 0 0 OK NR_024540 - - NR_024540 - - chr1:14361-29370 - - 15114.7 9501.21 20728.1 OK

Here are top 5 rows from output for yet another of the six samples:

tracking_id class_code nearest_ref_id gene_id gene_short_name tss_id locus length coverage FPKM FPKM_conf_lo FPKM_conf_hi FPKM_status NR_026820 - - NR_026820 - - chr1:34610-36081 - - 0 0 0 OK NM_001005484 - - NM_001005484 - - chr1:69090-70008 - - 0 0 0 OK NM_001005221 - - NM_001005221 - - chr1:367658-368597 - - 0 0 0 OK NR_046018 - - NR_046018 - - chr1:11873-14408 - - 0 0 0 OK NR_024540 - - NR_024540 - - chr1:14361-29370 - - 28376.8 20008.9 36744.6 OK

As a user, when selecting the same Reference Annotation to be used. I would expect the index of results by tracking id to be the same.

Why is this happening? Why are they not consistent?

If were to concat the above results across samples without correcting the indices, my data would be all mixed up.

I bring this to your attention as someone may not correct the indices before concat and therefore would be doing downstream analysis incorrect concatted data.

In turn, I would recommend that the index of results be corrected to be constant for a Reference Annotation when selected.

Thank you in advance for this consideration.

Best,

Kory Johnson

--------------------------------------------

Kory R. Johnson, MS, PhD Sr. Bioinformatics Scientist

<image001.jpg>

www.kellygovernmentsolutions.com

Providing Contract Services For:

Bioinformatics Section, Information Technology & Bioinformatics Program, Division of Intramural Research (DIR), National Institute of Neurological Disorders & Stroke (NINDS), National Institutes of Health (NIH), Bethesda, Maryland

Mailing Address:

NINDS/NIH Clinical Center (Building 10) Office 5S223 9000 Rockville Pike Bethesda, MD 20892

Contact Information:

Phone: 301-402-1956 Fax: 301-480-3563 email: johnsonko@ninds.nih.gov

P Green Message:

Please consider the environment before printing this e-mail. Thank you.

Important Message:

This electronic message transmission contains information intended for the recipient only. Such that, the information contained herein may be confidential, privaledged, or proprietary. If you are not the intended recipient, be aware that any disclosure, copying, distribution, or use of this information is strictly prohibited. If you have received this electronic information in error, please notify the sender immediately by telephone. Thank you.

--------------------------------------------

___________________________________________________________ Please keep all replies on the list by using "reply all" in your mail client. To manage your subscriptions to this and other Galaxy lists, please use the interface at:

http://lists.bx.psu.edu/