Question On Cufflinks Output

Question: Question On Cufflinks Output

7.4 years ago by

Wen Huang • 20 wrote:

Dear Galaxy team and users, I have a question on the output by cufflinks on Galaxy. I started with about 28M paired-end reads and mapped them to the reference genome using Tophat on Galaxy. The aligned fragments were assembled by cufflinks, again on Galaxy and I got an output with the first few lines on the bottom of this email. I was wondering how could cufflinks possibly estimate FPKM on the order of e+07 when the coverage is between 8-50 fragments per base and the total mapped fragments smaller than 28M. Assuming that 20M fragments were mapped, the FPKM should be something around coverage/28. Was the e in the output the Euler's number or 10? I appreciate your help. Thanks, Wen Huang tracking_id class_code nearest_ref_id gene_id gene_short_name tss_id locus length coverage status FPKM FPKM_conf_lo FPKM_conf_hi CUFF.2.1 - - CUFF.2 - - chr1:90301-90706 405 21.1837 OK 1.84527e+07 1.10716e+07 2.58338e+07 CUFF.1.1 - - CUFF.1 - - chr1:65419-65692 273 30.9833 OK 2.31848e+07 8.52143e+06 3.78481e+07 CUFF.3.1 - - CUFF.3 - - chr1:135255-135896 641 8.61389 OK 6.31907e+06 3.41968e+06 9.21846e+06 CUFF.4.1 - - CUFF.4 - - chr1:155808-156529 721 7.26147 OK 5.32695e+06 2.88278e+06 7.77112e+06 CUFF.5.1 - - CUFF.5 - - chr1:160421-160729 308 17.6004 OK 1.77483e+07 7.50132e+06 2.79953e+07 CUFF.6.1 - - CUFF.6 - - chr1:170695-171212 517 9.16414 OK 8.41605e+06 4.44869e+06 1.23834e+07 CUFF.7.1 - - CUFF.7 - - chr1:180885-181188 303 30.5702 OK 2.6515e+07 1.36533e+07 3.93767e+07 CUFF.8.1 - - CUFF.8 - - chr1:184397-184702 305 26.712 OK 2.13696e+07 9.94707e+06 3.27921e+07 CUFF.10.1 - - CUFF.10 - - chr1:233237-234095 858 3.71208 OK 3.31435e+06 1.60283e+06 5.02588e+06 CUFF.9.1 - - CUFF.9 - - chr1:203688-204070 382 41.6301 OK 5.36082e+07 4.02061e+07 6.70102e+07 CUFF.11.1 - - CUFF.11 - - chr1:239126-239664 538 19.5995 OK 2.0562e+07 1.45634e+07 2.65605e+07 CUFF.12.1 - - CUFF.12 - - chr1:243903-244327 424 10.3509 OK 1.07542e+07 5.37709e+06 1.61313e+07 CUFF.15.1 - - CUFF.15 - - chr1:240487-240995 508 15.8596 OK 1.83065e+07 1.23671e+07 2.42459e+07

rna-seq cufflinks • 2.0k views

ADD COMMENT • link •

modified 7.4 years ago by Jeremy Goecks • 2.2k • written 7.4 years ago by Wen Huang • 20

7.4 years ago by

Jeremy Goecks • 2.2k

Jeremy Goecks • 2.2k wrote:

Hello Wen, It's not necessary to send multiple emails to the mailing list; we track incoming emails to ensure that we respond to all of them. Your FPKM values do look high, but keep in mind that coverage is only part of the FPKM calculation; it's also dependent on transcript length and the total number of reads in your sample. Your transcript lengths look very short, so that may be skewing your FPKM values. For the record, Cufflinks is using scientific/E notation, so e denotes powers of 10 in the FPKM output. A good place to ask followup questions about cufflinks output is the cufflinks help email address: tophat.cufflinks@gmail.com Good luck, J.

ADD COMMENT • link written 7.4 years ago by Jeremy Goecks • 2.2k

Hi galaxy-users When trying to display a set of chip microarray data in UCSC main, , we encounter this error: Error(s): Error line 38879 of http://main.g2.bx.psu.edu/root/display_as?id=3221463&display_app=ucsc& authz_method=display_at: chromEnd larger than chrom chr21_random size (1679928 > 1679693) Can anybody help for this? Best, Yongde

ADD REPLY • link written 7.4 years ago by YBao • 50

Hello, The error is indicating that the end of your dataset is larger than the chromosome it is aligned to for this line (at least). This is expected when attempting display of certain data types at UCSC. This is a known issue. You can either remove these lines from your dataset with tools in Text Manipulation, download the data & trim coordinates using unix tools you design or obtain from UCSC, or best: view in Trackster ("Visualization"). Hopefully one of these solution will work for you, Best, Jen Galaxy team ps: For next time, it would be very helpful for us if you would start over in a completely fresh/new thread email when starting another discussion topic, to help us track and answer the most effectively. Thanks! -- Jennifer Jackson http://usegalaxy.org/ http://galaxyproject.org/

ADD REPLY • link written 7.4 years ago by Jennifer Hillman Jackson ♦ 25k

Similar posts • Search »