Dear all, I am trying to use CD-hit to remove the duplicates from the file that is the output from trinity (RNA seq assembly).
I used the following parameters:
cd-hit-est -i in.fasta -o out_cdhit90.fasta -c 0.90 -n 9 -d 0 -M 0 -T 0
But the output file still contains lots of small or fragmented sequence plus the best one. How can I remove those small or fragmented duplicates by changing the parameters?
thanks
ZQ
thanks, Mathew.
I want to remove the duplicates and the also some isoforms. My data was pooled RNAseq, so the diversity is in it.
If you just want to remove fragments below a certain length, you can use the argument. If you want to control via length of aligned vs unaligned lengths, there are various other arguments (e.g. -aL, -AL, -uL, -UL, -aS, -AS, -A) which allow you to manipulate this respective of the longer or shorter (or both) sequences in each pairwise comparison.
http://weizhongli-lab.org/lab-wiki/doku.php?id=cd-hit-user-guide#cd-hit-est
http://weizhongli-lab.org/lab-wiki/lib/exe/detail.php?id=cd-hit-user-guide&media=cd-hit-figure2.png
cd-hit-dup -i R1.fq -i2 R2.fq -o output-R1.fq -o2 output-R2.fq [other options] (for PE reads FASTQ)
cd-hit-dup -i input.fa -o output.fa [other options] (for single reads FASTQ)
Options: -i Input file (FASTQ or FASTA); -i2 Second input file (FASTQ or FASTA); -o Output file; -o2 Output file for R2; -d Description length (default 0, truncate at the first whitespace character) -u Length of prefix to be used in the analysis (default 0, for full/maximum length); -m Match length (true/false, default true); -e Maximum number/percent of mismatches allowed; -f Filter out chimeric clusters (true/false, default false); -s Minimum length of common sequence shared between a chimeric read and each of its parents (default 30, minimum 20); -a Abundance cutoff (default 1 without chimeric filtering, 2 with chimeric filtering); -b Abundance ratio between a parent read and a chimeric read (default 1); -p Dissimilarity control for chimeric filtering (default 1);
Hello, do you know any tool DIFFERENT from CD-hit to filter CDS unigenes.?
- Badges
- Science topic
- Similar topics
- Filing