cuffmerge fails with IOError - help

[adrian]

Hi:
Has any one faced this type of error.

I am unable to wrap my head around this error. I appreciate any suggestions or clues.

Thanks.

[Thu Feb 20 1:55:05 2013] Beginning transcriptome assembly merge

Traceback (most recent call last):
File "/temp/cufflinksVer2/cufflinks-2.0.2.Linux_x86_64/cuffmerge", line 576, in <module>
sys.exit(main())
File "/temp/cufflinksVer2/cufflinks-2.0.2.Linux_x86_64/cuffmerge", line 538, in main
gtf_input_files = test_input_files(transfrag_list_file)

File "/temp/cufflinksVer2/cufflinks-2.0.2.Linux_x86_64/cuffmerge", line 268, in test_input_files
g = open(line,"r")
IOError: [Errno 2] No such file or directory: 'chr1\tCufflinks\ttranscript\t62948\t63887\t1\t+\t.\tgene_id "ENSG00000240361"; transcript_id "ENST00000492842"; FPKM "0.0000000000"; frac "0.000000"; conf_lo "0.000000"; conf_hi "0.000000"; cov "0.000000";'

tophat --version
TopHat v1.2.0

cufflinks v2.0.2

merge_cuff_asms v1.0.0

[adrian]

apologies for posting in haste. It worked.
I made all gtf files in list and supplied that list.

[rz566]

Many people make the mistake of concatenating the content instead

From the Cufflink website, assemblies.txt should contain a list of file name:

~/xxx/transcripts.gtf
~/yyy/transcripts.gtf

Many people simply cat the content of transcripts.gtf into assemblies.txt which is wrong:

1\tCufflinks\ttranscript\t860106\t861365\t1000\t+\t.\tgene_id "CUFF.1"; transcript_id "CUFF.1.1"; FPKM "648.3693588401"; frac "1.000000"; conf_lo "140.350233"; conf_hi "269.243304"; cov "92.865564