Hi everyone, I was just wanting to ask about the trunq in DADA2.
is it better to use a high value such as 10-20 or the default value of 2?
Best wishes
Hello again Martyn,
As you have probably seen in the tutorials, DADA2 can also use the trunc and trim settings to cut out the low-quality sections of all reads, instead of cutting each read to a different length based on a Q score.
Both options can work well, but it can be easier to have reads of a consistent length, which you will not get using the --p-trunc-q setting.
If you want consistent read lengths or have paired-end reads, the setting shown in the tutorials are probably best:
--p-trim-left-f
--p-trim-left-r
--p-trunc-len-f
--p-trunc-len-r
If you have single-end reads and want to keep the most base pairs from each read, then the variable length truncation of --p-trunc-q could be nice. 
1 Like
Thank you again Colin , I see the default value is 2 in QIIME which seems low to me. In your opinion is it better to use a higher value?
Best wishes.
I don't use --p-trunc-q at all because it creates variable-length reads. Instead, I look at the read quality plots, choose a trimming location based on that quality, and trim and trunc all reads to the same length using the settings I mentioned above.
(I usually work with paired-end reads, so that's part of my decision. If you have only forward reads, say from the Ion Torrent platform, --p-trunc-q could be a good choice, especially with higher values like 10 or 20 like you mentioned!)
This topic was automatically closed 31 days after the last reply. New replies are no longer allowed.