This is a test version of Biostars. For the public version, visit https://www.biostars.org.
smallRNA aligner and pipeline

Hello :)

I have smallRNA sequencing data and I am looking for developing a pipeline. I want to analyze not only miRNA but also the other sncRNAs.

I have read some papers in which the authors have started with align the reads against known miRNA from miRBase; after that they took the unaligned reads and they aligned them against various sncRNA libraries.

Could you suggest an aligner for my aim? Do you think that is it a good choice to align before against miRNA database and after that against the others libraries?

Thank you in advance!

Best regards

smallrna pipeline aligner mirna
• 332 views • link
written by noeD noeD

thank you for your help. I have seen that posts but unfortunately I didn't find them useful :(

• 0 views • link
written by noeD noeD

1 answer

Many publications involving smallRNA use standard aligners (e.g. bowtie, BWA, etc..). In general, they align their reads to the genome and then subset their alignments based on metrics such as size and whether or not they match the locations of previously annotated RNA features.

The major obstacle is that many small RNA, especially siRNA and piRNA, align to repetitive regions. So using unique alignments is generally recommended.

• 0 views • link
written by benformatics benformatics

Thank you for your reply! I have a question: is it BWA able to align short reads like miRNA (~20n)?

Many publications involving smallRNA use standard aligners

Could you indicate that publications in order to understand how they handle the following steps for annotate the smallRNA. Until now I have found: "SePIA: RNA and small RNA sequence processing, integration, and analysis" and "iMir: An integrated pipeline for high-throughput analysis of small non-coding RNA data obtained by smallRNA-Seq".

Thank you in advance!

• 0 views • link
written by noeD noeD

If you use only unique genomic alignment, you would lose a lot of information. An alternative strategy would be to align all reads on the genome, then collapse repetitive RNAs together. Or, as the OP proposed, to first align to the annotated miRNAs/piRNAs/sncRNAs transcriptome then align the remaining reads to the genome.

• 0 views • link
written by Carlo Yague Carlo Yague

Sorry for the stupid question (I am new to smallRNA-sequencing) but:

Or, as the OP proposed, to first align to the annotated miRNAs/piRNAs/sncRNAs transcriptome then align the remaining reads to the genome

how is it possible to align in the same time against several transcriptome? Do you suggest to align before to miRNA and after that take the unalign reads and align them agains piRNAs and so on? An why align the remaining reads to genome?

Thank you for your help!

• 0 views • link
written by noeD noeD

how is it possible to align in the same time against several transcriptome? Do you suggest to align before to miRNA and after that take the unalign reads and align them agains piRNAs and so on?

I suggested to align to a transcriptome containing miRNAs, piRNAs and sncRNAs (and other RNAs if you want). So it is only one alignment.

why align the remaining reads to genome?

You could imagine that in your specie of interest, there are sRNAs expressed that are not annotated. Aligning the unmapped reads to the genome would allow you to (probably) see those unnanotated sRNAs. But you could also restrict your analysis to what is annotated. I guess it depends on the model and on the goal of the experiment.

• 0 views • link
written by Carlo Yague Carlo Yague

Thank you for your help, it is very useful for me! I have seen some pipeline in which the authors prefered to align the reads against miRNA transcriptome, and after that align the unmapped read against genome and annotate the others ncRNA (piRNA for example). Why? Are there benefit in that choice?

Thank you in advance

• 0 views • link
written by noeD noeD

I guess the reason is what I told you : without mapping to the genome, they would have ignored the 'new' ncRNAs in their analysis.

• 0 views • link
written by Carlo Yague Carlo Yague

Thank you (again) for you help. My question was another: why align before against miRNA transcriptome, after that against genome and annotate the others ncRNA rather than align against ncRNA transcriptome and after that take the unmapped for "discovering" new ncRNA not annotated. In the both cases you will obtain the reads that are not annotated...

• 0 views • link
written by noeD noeD

Sorry I misunderstood th question. I guess the reason is that it is faster and more efficient to align most reads to a non-redondant miRNA transcriptome than to the genome where there can be mapping ambiguities. The results of a transcritome mapping are also easier to analyse because it is straightforward to assign read counts to transcripts in this case.

• 0 views • link
written by Carlo Yague Carlo Yague

Log in to answer this question.

More posts like this