A reliable tool to predict attenuators in bacterial genomes

Question

[natasha.sernova]

Dear all,

I could miss the previous answer/explanation to the question, sorry for that!

I need a tool to predict attenuators in bacterial genomes.

I was told experienced people easily see them in any sequence, but

I am not an experienced person and I have 500 bacterial genomes…

Our home-made programs don’t find some known attenuators.

I found this article from 2010 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC2965389/pdf/gb-2010-11-9-r97.pdf) and

another one, from 2019 (https://www.ncbi.nlm.nih.gov/pmc/articles/PMC6524683/

). But is there something else, so far it's not enough?

Please, help me – I hope such a reliable soft has been already made.

Many thanks,

Natasha

Answer 1

[Mensur Dlakic]

From what I know about attenuators, they are cis-regulatory elements that are usually some kind of structured RNA. If so, you probably don't need new software to find them - you need good models. There is a suite of programs called Infernal that builds and scores covariance models that include both RNA sequence alignments and secondary structure features. It powers Rfam, which is a large database of regulatory RNAs.

As long as you can align and structurally annotate a good number of attenauator sequences to create a covariance model, Infernal should be able to find its matches even in the absence of high sequence identity.

[natasha.sernova]

Thank you very much, we'll try.

[Mensur Dlakic]

You have an example here and published here how Infernal is used to search for bacterial terminators.

[natasha.sernova]

That's a nice article. Plus 22 articles citing it. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3152330/citedby/ That's great, many thanks!

But they are mostly about terminators. Am I right and there are 2 possible roles for attenuators? What is the second one, or both are closely related to termination?

[Mensur Dlakic]

I don't know enough about attenuators to answer that question conclusively, but I am pretty sure that attenuators and terminators are related in terms of structure and function. The point is that you should be able to adopt their study with terminators to your problem, assuming you have enough attenuator sequences and can align them reliably by structural features.

Covariance models also score structural conservation with sequence conservation, which is why they are much better in scoring regulatory RNAs compared to sequence-based searches which find only close sequence homologs. If you look closely at the alignment of their second terminator, you will notice that many of them are not very similar in terms of sequence, and I suspect the same may be true for attenuators.

Main Rfam pages for terminators described in that paper are here and here.

[natasha.sernova]

Thank you very much!