Rna 'Gene' Prediction Tools

Question

[Andrea_Bio]

Hello

I would like to scan some sequence data for ncRNA and I'm a little bit at a loss where to start as there are simply too many tools to chose from.

The most established one seems to be NCRNASCAN.

Is there a preferred ncRNA scan tool?

Many thanks

[Andreas]

just saw the mirna tag. are you exclusively looking for mirnas?

[Andrea_Bio]

no, not specifically mirna. I thought i had added more tags than that.

Answer 1

[Lee Katz]

A combination between tRNA Scan SE and RNAmmer would be good. Then, if you have other ncRNAs to find beside rRNA and tRNA then you can use predictors for those too. I think it's too difficult to predict ncRNAs as a whole group; it should be much easier to predict one class of them at a time.

[Andrea_Bio]

i'm happy to do one groupa at a time. i want to scan an entire genome for each group for ab initio prediction but i can't find anything to do this, except for snoRNA

[Andrea_Bio]

sorry, of course there is rnammer as you suggested but i would also like to find mirna genes.

[Lee Katz]

Then, I suppose one of these packages would be good. http://www.bioinformatics.ca/links_directory/search/mirna

MicroInspector, RNAhybrid, ...

Answer 2

[Andreas]

There are a number of tools. The choice depends on the nature of your input.

Is it a single sequence or a set of homologous (and aligned) sequences? Is it a short sequence (ie. <1k) or a long, genomic sequence?

Also, do want to do de-novo prediction or look for homologous RNAs?

For starters have a look at

• "A practical guide to the art of RNA gene prediction".
• "From consensus structure prediction to RNA gene finding."
• Wikipedia's List of RNA structure prediction software

Andreas

[Andrea_Bio]

Thanks for answering. My input will be an entire genome for an organism. I would like to scan the genome to mine any potential RNA genes such as miRNA. The point behind this is I am looking for the consequences of SNPs in the genomes of 10-several hundred (don't know how many yet) individuals. I'm doing the typical ensembl consequence/polyphen route for SNPs in genes but it would be nice to try and say something about the SNPs in other areas too. I have just posted another question about this.