Hi all, I haven't been able to find an answer to this in the man pages nor the documentation- I'm running ABySS with paired end fragment reads, and ABySS is discarding about 25% of the reads as 'unchaste'. Can anyone tell me what criteria is used by ABySS for this filtering?
thanks!
Well, I found the answer- it's actually related to the 'chastity filter' on Illumina instruments. Reads marked as 'unchaste' have a higher noise level in the first 25 bases but are included in the fastq output; ABySS is apparently filtering those; but filtering can (should be?) suppressed with the --no-chastity option. http://support.illumina.com/help/SequencingAnalysisWorkflow/Content/Vault/Informatics/Sequencing_Analysis/CASAVA/swSEQ_mCA_PercentageofClustersP.htm Just in case anyone else comes searching for that answer!
Use of this site constitutes acceptance of our User Agreement and Privacy Policy.
version 2.3.6
If reads failing chastity filter are in your dataset then that is an unusual thing. Core facilities would generally never include these in datasets (they are excluded by default by Illumina basecalling software). Wonder if the facility you got the reads from was trying to inflate the output size by including those reads.
Hm, thanks for that suggestion! I'll have to check the other runs we've gotten from that facility. cheers- Scott