multiple mutations in tha same gene

[pg84]

Hi there,
I'm a postdoc in Boston (Longwood area). One of the projects I am developing includes sequencing (I am quite new in the field).
I have some sequencing data that is giving me problems. Has anybody seen before several mutations (2-15) in the same region (10-2000bp) in paired end reads? thay have good coverage, mapping quiality and are not repetitive sequences.
Any suggestion?
Thanks!!

[dpryan]

Do you mean you found them in individual reads (meaning that multiple reads covering the same region may not at all agree) or that you have multiple independent reads giving evidence for the polymorphisms? I've certainly seen that with Sanger sequencing of human DNA before. If you're looking at human DNA (rather than an inbred mouse strain or cell culture), you'll often find a number of SNPs.

[pg84]

Hi dpryan,
they are in multiple independent reads and the DNA comes from a clonal culture from cells of an inbred mouse strain (BL6).
thanks,

[GenoMax]

Can you comment on exact number of reads?

Are they all known to be of good quality (Q-scores) across the entire length of the read? Have the reads been trimmed?

[dpryan]

Hi pg84,

To continue along the same line as GenoMax's response, do the reads in this area seem to arise from PCR duplicates (i.e. a single round of crappy amplification could be causing the results you're seeing) or do you indeed have multiple reads starting from a variety of positions and with both orientations covering the positions?

Also, can you give a little more detail about the nature of the cell culture? If these are already differentiated cells that have been reprogrammed or otherwise immortalized, then that process could have screwed up some of the proof-reading machinery (though, one would expect increased polymorphisms globally in such cases).

[pg84]

Hi GenoMax,
Both the Qscore and the mapping quality are more than 30. I am not sure if the reads have been trimmed, I belive data was demultiplexed and aligned directly. We have checked some of those calls and they appear in the middle of the reads. and the number of independent reads are between 30 and 250.
thanks!

[pg84]

Hi dpryan,
duplicate reads were removed. we are loking at independent reads, they start in different positions and have differnet orientations. the amount of DNA was limiting, can that facilitate crappy PCRs? could that affect the independent reads?
About the culture, clones were grown from single primary cells that have not been affected in any way. Cultures were kept in low O2 in a rich media that supports their growth as happy as they could be.
I am looking for calls that appear in more than 30% of the reads, to look at initial events. So the polimorphism that could have arisen in the culture would be sorted out with the filters.
Thanks!

[GenoMax]

Hi dpryan,
duplicate reads were removed. we are loking at independent reads, they start in different positions and have differnet orientations.

Thanks!

You should perhaps put all ("duplicate") reads back in so you can have a more representative picture of which of the SNP's are more important than others. It is possible that you have a few reads with errors in them that are leading to a heterogeneous picture when looked at as a non-redundant set.

[pg84]

oh! I thought all duplicate reads should be removed

[GenoMax]

oh! I thought all duplicate reads should be removed

Perhaps not in this case. You could do that for ease of displaying against a reference but you will have to keep track of counts for each individual read that remains in the non-redundant set to give you an idea of the original distribution.

[pg84]

ok, I will try that.
thanks a lot!!