RNA-Seq or Exome Seq for Clinical data analysis

Question

[Karma]

My plan is to develop a tool that can predict gene mutation and gene's expression pattern in a user input sample (fastq). From literature survey, I have understood Exome seq is good for calling variants from sample as it call variants from all coding region and RNA-Seq is best for gene / transcript expression data analysis. But, people do go for variant analysis using RNA-seq also. In such case all genomic variants are not captured. Here I have three important question

1. Can I use RNAseq for identifying the variants for clinical analysis?
2. Is RNA-Seq clinically more important than exome seq as it is used for both variant calling and identifying transcript expression?
3. Are Mutations in expressed regions more important than mutations in non-expressed region? Please give your valuable suggestions.

If you need any clarification in the question please reply me here.

Sorry for my language.

Thank you

[GouthamAtla]

I think you have first understand what is exome seq and what is RNA-Seq, also some concepts like tissue specific expression patter, what is actually being sequenced in exome and RNA-Seq etc.

http://www.rna-seqblog.com/a-beginners-glance-at-rna-sequencing-vs-exome-sequencing/

There is also a review Exome versus transcriptome sequencing in identifying coding region variants

[Karma]

Thank you :) @geek_y I have seen this, but still confirming from experts like you. Do you have any suggestions for my 2 and 3 questions

[WouterDeCoster]

My apologies for being blunt, but I don't think you should be working on a clinical data analysis tools if you don't understand these basic concepts you are asking about. It's probably worth it to first read a lot about what you are planning to do.

[WouterDeCoster]

Do we need to tell you again about unnecessary upvotes of your comments/questions?

[Ram]

Yeah - I noticed something fishy too. Also check blast alignemnt query coverage - similar pattern of constant upvotes to trivial comments.

[WouterDeCoster]

After my comment 2 upvotes on the question itself were removed. Odd.

[Brian Bushnell]

I thought it was strange that I got +6 on a random answer that anyone could have given. In, like, 10 minutes.

[WouterDeCoster]

So, to summarize, we have

• fishy
• odd
• strange

That's rather remarkable, isn't it?

[Karma]

You got two upvotes for this trivial comment. Who did you arrange to upvote your comments? This is fishy, odd and stranage

[Ram]

In the current context of discussing user behavior, Wouter's comment is not trivial. I'd recommend some maturity in responses please, we are all scientists here.

[GouthamAtla]

I wondered about the same. :-p

Answer 1

[Brian Bushnell]

1) No.
2) No.  Variant-calling is less accurate from RNA-seq because there is no guarantee both chromosome copies are expressed (or that either is expressed at that time).
3) No.  Different genes are expressed at different times.  A gene that is only expressed, say, in embryonic development might be critically important but you'd never see it in adult RNA-seq.

So - you need to do WGS or exome-capture for mutations, and RNA-seq for expression.

[Karma]

Thank You.... Precise and Perfect answer

Answer 2

[GouthamAtla]

I am not an expert, but

1. RNA-Seq is useful to understand the tissue specific gene expression and for perturbation studies . BUT, if you want to do a clinical study, at most you would be doing RNA-Seq from saliva/blood and that might not be useful if the variant has effects the heart or pancreas , as you might not even have expression of those tissue specific genes in the blood/saliva sample.

2. Its been shown in several studies that the non-coding variants have large impact on gene expression, as they are usually enriched in regulatory regions like enhancers, promoters etc.