I want to compare my sequencing results with the reference gene sequence by Chinese population. Is there any database available online that can help me in downloading gene sequence? I have already compared sequencing results with the Ref. genes sequence available on NCBI, UCSC etc etc. The problem is, my control group is also showing almost similar mutations in about 7/12 genes I have screened. So I want to compare it with Chinese population genome sequence so that I can evaluate the reason.
Another question is, what is the best possible way to represent copy number variation data by whole genome sequencing? Any data analysis tool or suggestion?
Hi Madiha,
I don't know the answer to your first question. However, for copy number variation the baseline test is relative coverage depth. I would also recommend this pretty recent review article (Pirooznia, Goes and Zandi, Frontiers in Genetics 2015; 6:138):
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC4394692/
There are several ways, but due to what you are trying to do, I would heavily suggest The Exome Aggregation Consortium (ExAC) at http://exac.broadinstitute.org/ . This database is excellent and can tell you much more than other databases for ethnicity. However, these data are just SNV/SNPs/DNPs/etc. I do know some genes may have extra domains or repeats that are determined by ethnicity. To examine this will be hard and would require Sanger sequencing on a massive scale. So in short, no, you cannot determine copy number variation from next gen DNA seq results. Gap(s) in the sequence (where a copy number alteration has occurred) make it hard to align to the reference genome and at best, very long reads will only allow ~10 base pair gaps.
Thanks Ryan for the detailed answer. This database is indeed a great help. I already have data of copy number variation by whole genome sequencing but I am a bit confused how to utilize that data in a meaningful way and what possible experimental work I can design for that.
Great to hear, copy number information can be quite fun to work with as well. However, there are many different levels in which it is processed. Typically, the field likes to take the array used and run it through a Circular Binary Segmentation (CBS) algorithm (better than HMM, which also requires a training dataset) and then plug these data to make graphs and other downstream analysis, such as GISTIC. Much more information can probably be found through biostars. I cannot say to what degree your information has been processed, but if your data has segment mean scores, it is likely been processed through CBS. If this is the case, you can move forward with your data. If it is a matrix or an array format, you will need to process it. If you have CBS output, you can get genomic coordinates (ensure you are working with the correct reference genome e.g., hg18, 19, 38, etc) from the table browser at UCSC and convert the segment mean scores to copy number by taking (2^segmean_score)*2. This would be something to script and would be impossible by hand, unless you know your gene of interest. Might be best to understand what degree your data has been processed and then google what biologist friendly programs you can use it with.
- Badges
- Science topic
- Similar topics
- Computer Science and Engineering
- Bioinformatics
- Bioinformatic Tools