Tweet
Hi All
I am using the new workflow of:
Cufflinks -> Cuffmerge -> Cuffquant -> Cuffnorm
I am confused with the 'hits' parameters within cuffnorm:
CHOICE 1 --compatible-hits-norm: count hits compatible with reference RNAs only [ default: TRUE ]
CHOICE 2 --total-hits-norm: count all hits for normalization [ default: FALSE ]
First, I am assuming when they say 'reference RNAs' that is based on the annotation gtf you supply (which in my case is the output of cuffmerge, so the merge.gtf file) which would include 'novel' transcripts IF within cufflinks you chose --GTF-guide as the assembly option.
With this assumption , what would the disadvantage be to using Choice 2 ? Would their be regions within the transcriptome that are mapped YET not assigned as a novel transcript by cufflinks, so that they bias the normalization with cuffnorm? (if --total-hits-norm is ON)
Any clarity to the pros and cons of this parameter would be helpful!
I am using the new workflow of:
Cufflinks -> Cuffmerge -> Cuffquant -> Cuffnorm
I am confused with the 'hits' parameters within cuffnorm:
CHOICE 1 --compatible-hits-norm: count hits compatible with reference RNAs only [ default: TRUE ]
CHOICE 2 --total-hits-norm: count all hits for normalization [ default: FALSE ]
First, I am assuming when they say 'reference RNAs' that is based on the annotation gtf you supply (which in my case is the output of cuffmerge, so the merge.gtf file) which would include 'novel' transcripts IF within cufflinks you chose --GTF-guide as the assembly option.
With this assumption , what would the disadvantage be to using Choice 2 ? Would their be regions within the transcriptome that are mapped YET not assigned as a novel transcript by cufflinks, so that they bias the normalization with cuffnorm? (if --total-hits-norm is ON)
Any clarity to the pros and cons of this parameter would be helpful!