is there any database that show RNA false positive?

Question

[9521ljh]

i have RNA -seq file.

i run mutation calling

and Find particular gene is recurrently showing.

But, for more confidence, i want to identify whether this calling is false positive or not...

for checking, my idea is finding some database that record the false positive results.. but hard to find.

[Kevin Blighe]

How did you call these mutations? If you used GATK, then note the message on their website:

Finally, we know that the current recommended pipeline is producing both false positives (wrong variant calls) and false negatives (missed variants) errors. While some of those errors are inevitable in any pipeline, others are errors that we can and will address in future versions of the pipeline. A few examples of such errors are given in this article as well as our ideas for fixing them in the future.

[source: https://gatkforums.broadinstitute.org/gatk/discussion/3892/the-gatk-best-practices-for-variant-calling-on-rnaseq-in-full-detail]

To also link you up to my previous answer: A: Inferring genotype based on RNA sequnces

[WouterDeCoster]

Your question could use a lot more information and you should be much more specific about what you have, what you did and what you observe.

What does a false positive here mean for you? Do you think the variant is not actually really there or do you think it is real but not important?

Answer 1

[i.sudbery]

One obvious thing to try is to cross check against a database of known RNA-editing sites like REDIportal. However, such databases are envitably incomplete. I'd be extra cautious about any A:G change that is supported only by RNAseq.

You could also check if your gene has any pseduogenised copies in the genome.

Answer 2

[Kristoffer Vitting-Seerup]

Unfortunatly I have to say that I have not yet seen anybody who can with good accuracy call mutations from RNA so I would be very careful with interpretations. One of the most recent attempts I'm aware of is this paper - but they still have a false discovery rate of 29%. Using GATK directly will most likely give worse results.