[GenoMax]

It appears that your index files are not matching the reference files. That is likely the source of all the problems you have posted across several threads.

I would suggest getting the complete set of index, GTF and sequence files from one source so that they all match: http://cufflinks.cbcb.umd.edu/igenomes.html

[Charitra]

I have used all ensembl .. I downloaded the whole pack of tar.gz files from ensembl cos last time you recommended me... something is wrong... it is only at cuffmerge and cuffdiff.... I have tried both UCSC and Ensembl.
Well, I can it one more time and get back to you.
Thank you...

[GenoMax]

If you downloaded the "Ensembl" archive file from iGenomes then the genome file is located at /Homo_sapiens/Ensembl/GRCh37/Sequence/WholeGenomeFasta/genome.fa.

This genome file contains no "GL000*" fasta entries. So I am not sure where you are getting that error from if you are using this Ensembl genome file for cuffdiff analysis.

Can you post the command line you are using?

[Charitra]

Dear GenoMax
I thank you very much for your kind reply and willingness to help me out of this problem.

My commands and error are attached in the file named 'Errors_RNA-seq_Cuffmerge'. I also think that I didnt sorted .bam files, could that be a reason gettin errors ?
Please have a look and suggest me something...
Thank you..

If you downloaded the "Ensembl" archive file from iGenomes then the genome file is located at /Homo_sapiens/Ensembl/GRCh37/Sequence/WholeGenomeFasta/genome.fa.

This genome file contains no "GL000*" fasta entries. So I am not sure where you are getting that error from if you are using this Ensembl genome file for cuffdiff analysis.

Can you post the command line you are using?

[GenoMax]

Thanks for providing the detailed log file. You appear to be on right track.

In spite of those "Warnings" (they are not errors) are you getting output from the cuffmerge step?

I checked a few of the ID's in the "Warning" messages and most of those look to be pseudogenes or lincRNA's etc. Though they are present in the gtf file no fasta records are created for them when you use the "gffread" utility (http://cufflinks.cbcb.umd.edu/gff.html)

How much memory do you have on this machine? Are you using 32-bit or 64-bit ubuntu?

[Charitra]

I am using Server 64 bit ubuntu has enough memory for multiple tasks ~50GB..
your mail is a relief for me. I have visualised the outputs of cufflinks using cummeRbund... but I was not sure about my cuffmerge and cuffdiff...

Would you like to recommend me something please ?

Thanks for providing the detailed log file. You appear to be on right track.

In spite of those "Warnings" (they are not errors) are you getting output from the cuffmerge step?

I checked a few of the ID's in the "Warning" messages and most of those look to be pseudogenes or lincRNA's etc. Though they are present in the gtf file no fasta records are created for them when you use the "gffread" utility (http://cufflinks.cbcb.umd.edu/gff.html)

How much memory do you have on this machine? Are you using 32-bit or 64-bit ubuntu?

[GenoMax]

Are you getting output from Cuffmerge/Cuffdiff or there is nothing being produced?

[Charitra]

Results are there... I have not yet analysed properly... I just got the output from R but didnt read the results from R... results are not empty... there are genes but there is not significant differential gene expression but still not sure...
do you think.. i can continue with these warnings and analyse the R results ??

[Charitra]

Dear GenoMax

I got one doubt, if there are lincRNA's and I just ignore them, then I am trying to differentiate the expression level in my samples. Sometimes, these are needed for differentiation ? is that okay if I continue... or is there anything where I can modify my commands to get more accurate results.
Thanks

Thanks for providing the detailed log file. You appear to be on right track.

In spite of those "Warnings" (they are not errors) are you getting output from the cuffmerge step?

I checked a few of the ID's in the "Warning" messages and most of those look to be pseudogenes or lincRNA's etc. Though they are present in the gtf file no fasta records are created for them when you use the "gffread" utility (http://cufflinks.cbcb.umd.edu/gff.html)

How much memory do you have on this machine? Are you using 32-bit or 64-bit ubuntu?

[GenoMax]

Dear GenoMax

I got one doubt, if there are lincRNA's and I just ignore them, then I am trying to differentiate the expression level in my samples. Sometimes, these are needed for differentiation ? is that okay if I continue... or is there anything where I can modify my commands to get more accurate results.
Thanks

You should create a separate posting with this question since I can't answer your question (and biologists on the forum may not read this thread). With a new thread you will have a better chance of getting an answer.

[Charitra]

You should create a separate posting with this question since I can't answer your question (and biologists on the forum may not read this thread). With a new thread you will have a better chance of getting an answer.

Dear GenoMax,
I got cuffdiff results and , i think, gene name or somthing went wrong which was giving warnings because i can not fined my gene in cummeRbund but gene is there in database when i checked opening in excel. is this problem with cuffdiff ?

kind help appreciated..