Cuffdiff short name

[RockChalkJayhawk]

I tired running cuffdiff on two RNA samples: Liver & Intestine.
After aligning with tophat, I ran cuffcompare

cuffcompare -o GFF -r ../hg19.RefFlat.GFF3 ../Liver/transcripts.gtf ../Lung/transcripts.gtf

cuffdiff GFF.combined.gtf ../Liver/accepted_hits.sam ../Lung/accepted_hits.sam

but my genes file looks like this:

ref_trans_id	class_code	gene_short_name	tss_id	locus	q0_FPKM	q0_conf_lo	q0_conf_hi	q1_FPKM	q1_conf_lo	q1_conf_hi	ref_id
XLOC_000001-[chr22:17517459-17539682]	-	-	-	chr22:17517459-17539682	0	0	0	0.56927	0	1.37434	-
XLOC_000002-[chr22:17565848-17591387]	-	-	-	chr22:17565848-17591387	2.37959	0.696966	4.06222	15.9232	12.1168	19.7296	-
XLOC_000003-[chr22:17956627-18033845]	-	-	-	chr22:17956627-18033845	0.540063	0	1.16367	0.413014	0	0.889921	-
XLOC_000004-[chr22:18043182-18073647]	-	-	-	chr22:18043182-18073647	92.8549	79.4174	106.292	1.11535	0	2.40326	-
XLOC_000005-[chr22:18121484-18211987]	-	-	-	chr22:18121484-18211987	2.41355	0.706914	4.12019	1.38433	0.254029	2.51463	-
XLOC_000006-[chr22:18560685-18572206]	-	-	-	chr22:18560685-18572206	3.25714	0.953993	5.56028	3.11362	1.1444	5.08285	-
XLOC_000007-[chr22:18593558-18614498]	-	-	-	chr22:18593558-18614498	3.26743	0.704014	5.83085	0.960786	0	2.17638	-
XLOC_000008-[chr22:18632757-18660160]	-	-	-	chr22:18632757-18660160	3.48871	0.849702	6.12772	3.43404	1.14438	5.7237	-
XLOC_000009-[chr22:18893735-18899600]	-	-	-	chr22:18893735-18899600	25.8998	16.5909	35.2086	0	0	0	-

My transcripts* and genes files have the correct gene and transcript names:

gene_id	bundle_id	chr	left	right	FPKM
SAMD11	45228	chr1	861120	879961	0.794517
NOC2L	45228	chr1	879583	894679	25.374
ISG15	45228	chr1	948846	949915	3.06525
AGRN	45228	chr1	955502	991492	8.73339
C1orf159	45228	chr1	1017197	1051736	1.44667
SDF4	45228	chr1	1152288	1167447	89.863
UBE2J2	45228	chr1	1189293	1209234	12.5814
ACAP3	45228	chr1	1227763	1243269	5.66814
PUSL1	45228	chr1	1243993	1247056	2.45665

Any clues?

[DrD2009]

Sorry I can't really help you with your problem, but I'm trying to use Cufflinks and I see you have made progress where I have been failing and was wondering if you could help me.

I've been trying to supply a GFF file when I run cuffcompare, but the names are never assigned to the transcripts and they are all classified as class "u" or "."

brandon@brandon-desktop:~/arab/small$ cuffcompare -o 162_162E -r ~/Desktop/tair9_small_RNAs.gff 162/162.cufflinks/transcripts.gtf 162E/162E.cufflinks/transcripts.gtf 
Warning: found 79213 transcripts with undetermined strand.
Warning: found 64824 transcripts with undetermined strand.

Any idea what I could be doing wrong? This problem has been holding me up for a while.

[RockChalkJayhawk]

Sorry I can't really help you with your problem, but I'm trying to use Cufflinks and I see you have made progress where I have been failing and was wondering if you could help me.

I've been trying to supply a GFF file when I run cuffcompare, but the names are never assigned to the transcripts and they are all classified as class "u" or "."

brandon@brandon-desktop:~/arab/small$ cuffcompare -o 162_162E -r ~/Desktop/tair9_small_RNAs.gff 162/162.cufflinks/transcripts.gtf 162E/162E.cufflinks/transcripts.gtf 
Warning: found 79213 transcripts with undetermined strand.
Warning: found 64824 transcripts with undetermined strand.

Any idea what I could be doing wrong? This problem has been holding me up for a while.

You might check your reference file to see if the stand information is in the correct place.

I really wish there were some repository where we all could download the same files and make this process a lot easier!

[DrD2009]

Thanks. I'll double check the file.

I agree. One day it will be done, probably after I stop working on this stuff. lol

[DrD2009]

I was wondering if you could post a few lines of the GFF file you used to annotate your reads for cuffcompare?

I've tried GFF and GFF3, but the reads still came out without annotations.

Are you able to supply cufflinks with a GFF as well to provide annotation? The manual only mentions GTF files.

Thanks again,
Brandon

[RockChalkJayhawk]

I was wondering if you could post a few lines of the GFF file you used to annotate your reads for cuffcompare?

I've tried GFF and GFF3, but the reads still came out without annotations.

Are you able to supply cufflinks with a GFF as well to provide annotation? The manual only mentions GTF files.

Thanks again,
Brandon

They only take GTF. I can send you the GTF file if you like. this is how I run the analysis:

tophat -p 5 - SampleName Sample.fq

cufflinks -p 5 -L Sample1 -G hg19.Ens.GTF Sample1.sam

cuffcompare -o Sample1 -r hg19.Ens.GTF -R Sample1.transcripts.gtf Sample2.transcripts.gtf

cuffdiff -p 5 combined.gtf Sample1.sam Sample2.sam

The GTF looks like this:

chr1 protein_coding CDS 67050223 67050289 . + 1 gene_id "ENSG00000173020"; transcript_id "ENST00000308595"; exon_number "14"; gene_name "ADRBK1"; transcript_name "ADRBK1-201"; protein_id "ENSP00000312262";
chr1 protein_coding CDS 67050223 67050289 . + 1 gene_id "ENSG00000173020"; transcript_id "ENST00000416281"; exon_number "2"; gene_name "ADRBK1"; transcript_name "ADRBK1-202"; protein_id "ENSP00000407159";
chr1 protein_coding CDS 67050599 67050699 . + 0 gene_id "ENSG00000173020"; transcript_id "ENST00000308595"; exon_number "15"; gene_name "ADRBK1"; transcript_name "ADRBK1-201"; protein_id "ENSP00000312262";

I can send you the GTF if you like.

[DrD2009]

Ahh, then that is my problem. The organism I work on, Arabidopsis thaliana, has no published GTF files anywhere that I have been able to locate to provide annotation for Cufflinks.

I have only GFF files. I might try creating GTFs out of my GFFs and seeing if I can provide annotation with Cufflinks that way.

Thank you for all of your help and fast replies. I really appreciate it.

[mrfox]

isoforms with cufflinks and cuffdiff

Hi everyone,
I have questions on the rebuilt isofroms/transcripts in the different steps of Cufflinks tool. I found that the rebuilt transcripts recovered by cufflinks and the transcripts in isoform_exp.diff created by cuffdiff are different. My questions are:
1)Cuffdiff re-assign the reads, rebuild the transcripts, and re-estimate the expression level?
2)How can I track the structure of the transcripts in isoform_exp.diff?

Thank you.

[mrfox]

sorry i posted the above thread in the wrong place.