Haplotype phasing from Miseq 150bp paired end reads?

[michael-steffen]

I am looking for a way to do haplotype phasing on 2 different genes that I have extracted using 150bp paired end reads on a Miseq. One gene is around 250bp the other around 600bp. This data is going to be used to supplement some data that was derived from traditional sequencing and cloning techniques. So, essentially what I am trying to do is replicate cloning but using illumina sequences instead. Is there any good advice out there onto how to do this, or if it is even possible?

Also, I posted a similar but different thread yesterday. I tried to figure out how to delete it but didn't see an option any where. Any help with this also?

Thanks,
Mike

[sarvidsson]

Did you map the data and had a look at it (with e.g. IGV)? Variants in such short genes could be phased by hand if the fragment size is large enough (read length isn't important, fragment size is). You might get problems with variants on the extreme edges of the genes, but linkage should be high with such short distances.

[michael-steffen]

I'm going to give this a try right now. It is something I tried to do in CLC, but it was difficult and left me with some doubts. You are right, the viewer is difficult.

I am trying to figure out at the moment how to import mapped reads into IGV. What do you typically do?

Thanks,
Mike

[sarvidsson]

I don't use CLC to align reads, but you can export its alignments in the SAM format, which then needs to be converted to BAM and sorted (search the forum for commands, there'll be plenty of examples). BAM can be opened directly in IGV and Tablet, along with a matching genome/FASTA.

I'd align the reads to the clipped gene region (i.e. a each gene sequence alone) with another mapper though (e.g. GSNAP is quite nice for such use) and load the results in Tablet (if you only have one sample, for several samples I'd use IGV) - the graphical representation of the linked pairs is good in Tablet.