Welcome to the New Seqanswers!

Welcome to the new Seqanswers! We'd love your feedback, please post any you have to this topic: New Seqanswers Feedback.
See more
See less

Cufflink's handling of reads

  • Filter
  • Time
  • Show
Clear All
new posts

  • Cufflink's handling of reads

    I am using Cufflinks to analyze RNAseq data and comparing different samples.
    I run the programs with the following options:
    cufflinks --num-threads 5 --GTF $ref_transcriptome --verbose --output-dir lane5/ $gene_name.lane5.bam >cufflinks.lane5.log 2>&1
    cuffcompare -o summary.stats -r $ref_transcriptome -V lane5/transcripts.gtf lane6/transcripts.gtf
    1) How does cufflink deal with reads only partially overlapping an exon (from a GTF file)?
    2) How are reads administered between overlapping genes with overlapping exons (the RNA data are unstranded)?
    3) If a read is only partially overlapping with the exon, will that part contribute to the FPKM value?
    4) How is the minimum number of reads needed in a transcript to produce an FPKM value controlled? I can often visually se reads overlapping a transcript in my GTF file, but still giving scores of zero in the transcripts.expr