De - multiplexing

[vl80]

We have incorporated unique tags/indexes/barcodes (5nt long) to our inhouse PCR primers. After PCR,amplicons from 8 samples were cleaned and pooled together in an equimolar manner and we have 10 such "pooledsamples". Then each of the "pooled samples" were prepared using TruSeq and run on Miseq. So, in the end, they "pooled samples" would have an illumina index each and each of the "illumina indexes" would have reads originating from 8 samples. Now, we need to identify a method to decode the samples properly. I must say, I am a total "dummy" when it comes to bioinformatics. Approaches to this and any software suggestions are welcome.

[JackieBadger]

You can use simple unix functions such as grep ...
or use jMHC http://code.google.com/p/jmhc/
or

FlexBar http://seqanswers.com/wiki/Flexbar

Software list: http://seqanswers.com/wiki/

[vl80]

Thank you very much. Just a few "dummy" questions..
The file that we get from Miseq after decoding of the illumima index is in the format of a fastaQ file. As a clinician who is at complete loss when it comes to bioinformatics and running scripts, the programme that I am looking for is something into which we can input the fastQ files and then input the nt sequence of the codes, preferably on a webinterface and which would give the decoded format in preferably a fasta file. jMHC seems to fit the criteria a lot however, their input requirement is a fasta file. Has anyone tried jMHC for data other than MHC related and using fastQ to fasta converted files? Thank you..

[JackieBadger]

convert using Galaxy web portal https://main.g2.bx.psu.edu

However..... If you are to embark on next generation sequencing data analysis there really is no way around knowing the basics of Unix/Linux and command line programs.
These are much more powerful that GUIs and can make your life MUCH MUCH easier.
A good place to start :http://korflab.ucdavis.edu/Unix_and_Perl/index.html

For example: jMHC ~ around 10 times faster on a Mac than PC...I have no idea why. But when I open jMHC on my Mac I do so from the terminal. This is so I can allocate more memory to Java (which jMHC runs in) so it can process my large data sets.
I do this by typing :

$ cd Desktop/jMHC
$ java -Xmx4g -jar jMHC.jar

This opens jMHC using 4gb RAM. If I do not do this jMHC will either freeze or crash on large data sets ~1M reads

It does not matter what amplicon you want to de-multiplex. jMHC was named so just because the group work on MHC.