small rna length distribution

Question

[bioinfo_ga]

I am performing small rna analysis and using srna-workbench for data trimming (Adaptor trimming and length filtering). After length trimming (16-40 bases) i am observing the distict reads are maximum of 32 bp whereas for small rna peak is observed at 18-22bp. What could be the reason?

[Titus]

Hi ,

How do your performed your trimming step with list of adaptor sequence ?

What tool do you used ?

Are you sure you can t have 32 bp RNA possible ? I was thinking in Piwi RNA for exemple .

Best

[bioinfo_ga]

I have used small rna workbench tool and used inbuilt adapter (LMN_3). I went through published papers of which almost all have reported the same trend (18-22bp). I checked for piwi RNA also but it is <0.02% in my data.

[Titus]

ok may be just check with fastq if you have a correct nuclear acid repartition in your trimmed reads.

[benformatics]

You would not expect very much piRNA unless you are deriving your sequencing samples from the core reproductive organs.*

*Assuming you aren't researching plants

[weixuchu63]

Hello bioinfo_ga,

What's your experience with srna-workbench? I am trying to analyze my plant srna seq data with it. But I don't know how to install it in my windows system :(. Could you share your experience with me? Appreciate sincerely.

Wei Xu

Answer 1

[benformatics]

Depending on the protocol used (which you did not mention so it is hard to tell) you could have degradation products remaining from turn-over of endogenous mRNA.

However, a smallRNA peak at 18-22bp makes sense... at least in animals as this would correspond to high levels of miRNA and/or siRNA. Generally smallRNA have the following ballpark size ranges (you should investigate what is expected for your organism of interest):

• 21-22 nt siRNA
• 21-23 nt miRNA
• 24-31 nt piRNA

The smaller sizes could be the result of over-trimming, the mRNA byproducts as first mentioned, and/or minor degradation of your small reads. If the distribution of your read lengths (i.e. molecule lengths) matches what has been previously observed in the literature I wouldn't worry too much.

Here is an example for Drosophila from this paper:

Answer 2

[davmlaw]

Hi, as well as plotting the length, you can also check the biotype (ie species of RNA) of overlapping genes

I've written a Python library called PyReference that includes a command to do this:

# Install
python3 -m pip install pyreference
# Process GTF into a JSON file
pyreference_gff_to_json.py --gtf gencode.hg38.gtf
# Create a setup file (see website for details)
# Run against your BAM file
pyreference_biotype.py ${BAM_FILE}

This generates a graph:

As well as a CSV of read counts per biotype for each length