Folks,

I have tried  pasting several GTF files (my merged gtf from 
cufflinks/cuffmerge) into the UCSC Genome browser.

I can add some header lines to the GTF file, eg:

browser position chr2:1-20000
browser full 'Assembly' 'Ensembl Genes' 'RefSeq Genes' 'miRNA'
track name=Cufflinks description=merged.gtf

then the regular gtf data.
Rat genome  annotations seem to refer to chromosomes as "1" , "2" etc  
not "chr1" ,"chr2" etc.
I fixed all the  chromosome names to  include the "chr" prefix with an 
awk script.

The header lines are accepted by the UCSC Genome browser  but so far I 
keep getting  this error off the UCSC browser when the reader gets to 
the first line of data

*Error File 'chrxaa_test.txt' - Unrecognized format line 4 of custom 
track: chr1 Cufflinks exon 12821949 12822044 . + . gene_id XLOC_000055; 
transcript_id TCONS_00000116; exon_number 1; gene_name Txlnb; oId 
ENSRNOT00000016553; nearest_ref ENSRNOT00000016553; class_code =; tss_id 
TSS82; p_id P50; (note: chrom names are case sensitive)

*
I want  to use**the merged.gtf file since it contains the TCONS_###  
and  coordinates.

I have tried adding header lines (browser and track lines), amending the 
chromosome names, converting all spaces to tabs and just loading the 
un-modified  GTF from cuffmerge. All efforts result in the first line of 
data confusing the UCSC browser.

The GTF format is listed as one that UCSC Genome Browser can read and 
use for a custom track but so far its not working for me. The first 8 
fields are standard but the 9th field have a variable format .

  I just spent about half an hour converting 
(https://lists.soe.ucsc.edu/pipermail/genome/2011-April/025696.html) my 
merged GTF to BigBed format to see if that worked but in the end I had 
lost the TCONS   labels which is what I want to try to display...

Any suggestions?

Starr
_______________________________________________
Genome maillist  -  [email protected]
https://lists.soe.ucsc.edu/mailman/listinfo/genome

Reply via email to