[myRNA_Seq]

cufflinks warning

I downloaded latest version of cufflinks binary and also added them to my path as following:
export PATH=$PATH:~/cufflinks-2.0.2.Linux_x86_64/cufflinks
export PATH=$PATH:~/cufflinks-2.0.2.Linux_x86_64/cuffdiff
export PATH=$PATH:~/cufflinks-2.0.2.Linux_x86_64/cuffcompare

When I tried to run cuffcompare, it gave me a warning:

Warning: Your version of Cufflinks is not up-to-date. It is recommended that you upgrade to Cufflinks v2.0.2 to benefit from the most recent features and bug fixes (http://cufflinks.cbcb.umd.edu).

and without *.tmap file generated.

Is there any problem in terms of the path setting?

Thanks a lot

[brdido]

myRNA_Seq, i would check if there is a copy of cufflinks in other PATHs.

i.e.: /usr/bin /usr/local/bin

[myRNA_Seq]

Thank you Brdido. I have just checked. There isn't a copy of cufflinks in other PATHs.

[brdido]

Have you tried to run cuffcompare using its full path?

~/cufflinks-2.0.2.Linux_x86_64/cuffcompare

If the message doesn't appear in the example above means that cuffcompare is healthy! I would suggest you to search all of the computer for cuffcompare (using find).

And remember that the variable PATH will be updated when you start a new terminal.

Other thing to try is to see if older cuffcompare is in your PATH:

echo $PATH

Any clues from here?

good luck! But this is more a linux question than a cufflinks question, maybe linux gurus can help you better than me...

[myRNA_Seq]

Than you Brdido. I have tried full path. It works without warning anymore, and got all the output files except *.refmap *.tmap file. Do you know why?

I followed the instruction from the paper "Differential gene and transcript expression analysis of RNA-seq experiments with TopHat and Cufflinks" (http://www.nature.com/nprot/journal/....2012.016.html) and use exact files in the examples.

$ find . -name transcripts.gtf > gtf_out_list.txt
$ ~/cufflinks-2.0.2.Linux_x86_64/cuffcompare - i gtf_out_list.txt -r gene.gtf

[brdido]

Have you checked if the chromosomes names are identical in transcripts.gtf and gene.gtf?

It is case sensitive AND some databases the id is just the number of the chromosome. For instance chromosome 1:

hg19 : chr1
b37: 1

[myRNA_Seq]

No. Not all the chromosomes names are identical between transcripts.gtf and gene.gtf.

[brdido]

So, have you corrected?

Did it worked?

[myRNA_Seq]

No, it still doesn't work.

I used full path to run cufflinks:
$ ~/cufflinks-2.0.2.Linux_x86_64/cuffcompare - i gtf_out_list.txt -r gene.gtf

No error/warning messages, I still can not get the output file *.refmap *.tmap files.

[myRNA_Seq]

Thank you very much brdido. It works for me now. I re-run some of the protocol to get the searchable index for each map files. it could be because of some of the program in wrong path and the old one was working there.

[xhb10108]

gpertea is Geo Pertea. He wrote cuffcompare, and he and I came up with these classification codes and the rules by which they are assigned to fragments while we were designing the program together.

Hi Cole Trapnell;

I've got a problem with cuffcompare.
I use a transcript compare with the reference.gtf from the ensembl web using cuffcompare,there are 31870 transcripts in my cufflinks output files,but after cuffcompare step,i check the tracking file,find that only 31340 transcripts are list in the file(taht is say those transcript are compared to the reference gtf file).
if i use the combinded.gtf file to compared to the reference ,the less transcript list in the trancking file.

I don't know why! It seems that cuffcompare eliminate transcripts in the creation of the new gtf file.

[delasa]

gpertea is Geo Pertea. He wrote cuffcompare, and he and I came up with these classification codes and the rules by which they are assigned to fragments while we were designing the program together.

Hi Dr.Trapnell,
I have a question regarding to the cuffcompare classes. Does class "j" also include the novel isoforms which may have intron retention (class i)? I mean an isoform may also fall into new splice junction and also have intron retention, but it is classified as class "j".

[BioCathy]

Hi Cole Trapnell;

I've got a problem with cuffcompare.
I use a transcript compare with the reference.gtf from the ensembl web using cuffcompare,there are 31870 transcripts in my cufflinks output files,but after cuffcompare step,i check the tracking file,find that only 31340 transcripts are list in the file(taht is say those transcript are compared to the reference gtf file).
if i use the combinded.gtf file to compared to the reference ,the less transcript list in the trancking file.

I don't know why! It seems that cuffcompare eliminate transcripts in the creation of the new gtf file.

Hi xhb10108,

Do you find the answers? I am also confused with that now. Could you share your opinion with me please?

Thank you so much!