Tweet
Dear all,
I have a question regarding running Cuffmerge on Galaxy. Any help will be appreciated!
My goal is to analyze for differential gene expression in a bacterium, R. centenum, in 5 timepoints (3 replicates for each timepoint). R. centenum has about 4000 genes. My workflow is: Bowtie- Cufflink - Cuffmerge - Cuffdiff. I used this workflow for my last data analysis and it worked great.
Now I am tried to do the same for my new set of data. After Cufflink, everything seems normal: cufflinks files have around 4000 lines, FPKMs good values and FPKM status are mostly OK. But when I tried to Cuffmerge two Cufflinks files together, the Cuffmerged file has only 48 lines. When I tried to Cuffmerge other Cufflinks files, no matter which cufflinks file for input (I tried around 8 different cufflinks files), the output Cuffmerge file are exactly the same with only 48 lines (which supposed to be 4000 lines).
However if I use Cuffcompare instead of Cuffmerge, the output file has 4000 lines.
I understand that Cuffmerge runs a Reference Annotation Based Transcript assembler (RABT) but Cuffcompare does not. So my question is, does this mean my data set is bad or my reference annotation is bad?
And also, should I use Tophat instead of Bowtie? Would it help?
The parameter I used for Cufflinks are:
Max Intron Length 1000
Min Isoform Fraction 0.1
Pre MRNA Fraction 0.15
Perform quartile normalization No
Use Reference Annotation Use reference annotation
Reference Annotation 28: R.centenum_annotation_su.gtf
Perform Bias Correction No
Set Parameters for Paired-end Reads? (not recommended) No
Global model (for use in Trackster) No dataset
How long will your job need 1 hr
The parameters I used for Cuffmerge are:
GTF file produced by Cufflinks 76: 26 1 CENS Cufflinks on data 73 and data 28: assembled transcripts
GTF file produced by Cufflinks 103: 26 1 D2 Cufflinks on data 100 and data 28: assembled transcripts
Use Reference Annotation Yes
Reference Annotation 28: R.centenum_annotation_su.gtf
Use Sequence Data Yes
Choose the source for the reference list history
Using reference file 27: R. centenum genome sequence
Thank you so much,
Qian
I have a question regarding running Cuffmerge on Galaxy. Any help will be appreciated!
My goal is to analyze for differential gene expression in a bacterium, R. centenum, in 5 timepoints (3 replicates for each timepoint). R. centenum has about 4000 genes. My workflow is: Bowtie- Cufflink - Cuffmerge - Cuffdiff. I used this workflow for my last data analysis and it worked great.
Now I am tried to do the same for my new set of data. After Cufflink, everything seems normal: cufflinks files have around 4000 lines, FPKMs good values and FPKM status are mostly OK. But when I tried to Cuffmerge two Cufflinks files together, the Cuffmerged file has only 48 lines. When I tried to Cuffmerge other Cufflinks files, no matter which cufflinks file for input (I tried around 8 different cufflinks files), the output Cuffmerge file are exactly the same with only 48 lines (which supposed to be 4000 lines).
However if I use Cuffcompare instead of Cuffmerge, the output file has 4000 lines.
I understand that Cuffmerge runs a Reference Annotation Based Transcript assembler (RABT) but Cuffcompare does not. So my question is, does this mean my data set is bad or my reference annotation is bad?
And also, should I use Tophat instead of Bowtie? Would it help?
The parameter I used for Cufflinks are:
Max Intron Length 1000
Min Isoform Fraction 0.1
Pre MRNA Fraction 0.15
Perform quartile normalization No
Use Reference Annotation Use reference annotation
Reference Annotation 28: R.centenum_annotation_su.gtf
Perform Bias Correction No
Set Parameters for Paired-end Reads? (not recommended) No
Global model (for use in Trackster) No dataset
How long will your job need 1 hr
The parameters I used for Cuffmerge are:
GTF file produced by Cufflinks 76: 26 1 CENS Cufflinks on data 73 and data 28: assembled transcripts
GTF file produced by Cufflinks 103: 26 1 D2 Cufflinks on data 100 and data 28: assembled transcripts
Use Reference Annotation Yes
Reference Annotation 28: R.centenum_annotation_su.gtf
Use Sequence Data Yes
Choose the source for the reference list history
Using reference file 27: R. centenum genome sequence
Thank you so much,
Qian