Fastx-tool kit_setting the prarameter

[chayan]

Hii
I am new to this. So can any one please suggest me how to set the parameter in fastx-tool kit regardgin the follwing criteria ....selection of the reads will be based upon Q20 value where min read length will be >100bp at a strech..if its find low quality reads at the 5'/3' end it will simply trim those portion and will take the rest of the read.... i tried
$fastq_quality_filter -q 20 and then
$fastx_clipper -a 0 -l 100 (-a 0; as the sequences are already clipped for the adapter as well as barcodes)
it is something like, i want trim the reads based on their quality value..is it possible with Fastx-tool kit at all??

Thanx for any help in advance

Regards

[mgogol]

I don't think fastx can do that. You may try trimmomatic? Or some other options here:

http://www.biostars.org/p/53528/

[chayan]

i have used Trimmomatics with the following parameter...

$java -classpath path/Trimmomatic-0.32/trimmomatic-0.32.jar org.usadellab.trimmomatic.TrimmomaticSE -phred33 input.fastq output.fastq SLIDINGWINDOW:100:20 MINLEN:100

it seems all ok..but at this point i want to be sure that does this processed the reads as i wanted..selection of the reads will be based upon Q20 value where min read length will be >100bp at a strech and this sliding window of 100 will make sure that if a read contain bases with lower qualities at both the end, it will simply trim those bases and keep the rest part of the read if its more than 100bp..does the parameters did the same thing?? am confused..any suggesition would be much appreciated........

[mastal]

For trimming low quality bases from the 5' end or the 3' end of the read, trimmomatic also has the commands LEADING or TRAILING, respectively.

Read the trimmomatic web page.

http://www.usadellab.org/cms/?page=trimmomatic

In addition, if you create a logfile when you run trimmomatic, it will tell you where it trimmed each read, so you can check a few of the reads to see if trimmomatic is doing what you want it to do.

[chayan]

dear mastal
Thanks a lot, yes i have seen that too but as i can see from the stats that my reads are of mix population, so basically i cant set a fix point of trimming at the 5'/3' position, in contrast 100bp with Q20, i guess, is enough stringency added for any kind of down stream analysis.. here my target is to OTU analysis.. can you please suggest me how to decode the log file..i found it really hard to made any conclusion about the trimming position

Again thaks for your kind help

[mastal]

See the trimmomatic web page, the section with the heading Single End Mode, for an explanation of the fields in the log file.

With the TRAILING, LEADING commands you don't specify how many bases or at which position to cut, you specify the base quality threshold, so for example TRAILING:3, will trim all bases with quality less than 3 (Ns, for example) from the 3' end of a read.

[chayan]

okk thanx again..now what will be the case if i choose LEADING:20 (as reads are in that direction), will it stop checking whenever it reaches to a bases above that specified quality value or it will continue its search and will look for the MINLEN:100 before taking the decision wheather to keep the read or throw it, because for obvious some bases will be of low quality at the other end of the read???

[chayan]

well, from the example section i have found that if i set the parameter like this SLIDINGWINDOW:<1 to 5>:20 (in eg. 4) and MINLEN:100 should solve my purpose, ...because setting SLIDINGWINDOW:100:20 will lead to loss of too many reads and not the true reflection of the quality of the run...let me try first and i will inform as things turns out...... many thanks for your help