Clare86
I was looking at various genes in the dna explorer and noticed that exon1 of SHANK3 has almost no reads. Does anyone else have the same thing in their data?
Screenshot 2020-06-28 at 14.55.08.png 
Quote 0 0
Erik Chapman
Is there a manual somewhere for how to use the explorer?  It doesn't seem intuitive. 

It's common for genome sequencing to be missing around 5% where there are gaps in coverage.  I don't know if that has to do with the process, or problems with the specific gene in question.  I think one common issue is where the bases have repetitive patterns that are long in comparison to the short read length, which can happen after the ends of genes, the result can be unclear. 
Quote 0 0
Erik Chapman
It sounds like Fullgenomes has an a possible antidote to that issue with their linked reads, but that is really expensive.  
Quote 0 0
Clare86
I know some regions are hard to sequence which is why I was curious if anyone else saw the same thing in their data. I got the same result from my Sano Genetics and Nebula Genomics tests so it's not just Dante's data.

The explorer is an implementation of IGV. There is a website about the desktop version which includes a user guide,  http://software.broadinstitute.org/software/igv/UserGuide.

The fullgenomes test is rather expensive. I'm not sure their linked reads test would help. They are still doing short reads just marking the ends of them so they can be stitched back together. I'm hoping that as long read sequencing is used more somewhere will offer it as a dtc test.
Quote 0 0
Umesh Reclusa hominid indica
shank 3.jpg Mine is in a similar/worse condition. However Clinvar doesn't list any documented highly clinically significant mutation in this region. BTW could you tell more about  interpretation of genomic data displayed on IGV, specifically the colors, I had uploaded a post in my earlier thread, highlighting different types of colors. If you could kindly share what your understanding of those is.
Quote 0 0
Umesh Reclusa hominid indica
shank 3.jpg Zoomed in View. Some loci have been sampled only 4 times.
Quote 0 0
Clare86
Thanks that's helpful. Suggests the region is simply hard to sequence. Although clinvar doesn't currently contain any references to deletions of just exon 1 of shank3 I would imagine the loss of this exon would be fairly significant and potentially cause Phelan Mcdermid Syndrome which is the primary condition associated with shank3.
Quote 1 0
Clare86
What about this region of shank3, chr22:51,135,585-51,136,531? Do you get less reads there too?
Screenshot 2020-07-07 at 12.15.13.png 
Quote 0 0
Umesh Reclusa hominid indica
Yup, mine even has an area with 0 reads!!  shank 3a.jpg 
Quote 0 0
Randy H
This is more often an issue of aligning short reads to create the BAM.  The DNA is sequenced.  It is just short sequences look similar to ,ultiple regions. The final assembly file (human genome reference model) will prioritize which area of the genome the short reads that are similar get mapped too.  Only reads that overlap the non-common edges enough will be uniquely and correctly mapped. Large CNV and SV areas will have this issue.  The only solution is longer reads that can overlap the "common" regions fully.
Quote 0 0