miRNA Quantifiaction Tool

Question

[anthony.santana.703.j]

So im struggling to find a tool to quantify the reads obtained from the Qiagen mirna library prep kit. Most places online tell me to use an already established pipleline to quanitfy mirna, but they have their own pipelines and settings. Im working on developing my own pipeline and ive read the featurecounts may have an issue when aligned to mirna, the settings could be adjusted to be more cautious of this. But imm left wondering if featurecounts is sufficient enough to quantify known mirna. Other piplines dedup but dont do it based on umi, which is already part of my pipeline.

Answer 1

[GenoMax]

Im working on developing my own pipeline

Is there a specific reason you want to do that? Even if you are learning the process, there are only so many ways the data can be analyzed. A pre-existing pipeline can save time and effort.

But imm left wondering if featurecounts is sufficient enough to quantify known mirna.

featureCounts should be fully sufficient to do the counting as long as you have a properly aligned BAM along with the right annotation. Post your command lines if you are running into specific issues.

There is a prior post for additional reference: miRNA low mapping(Qiagen miRNA Library Prep) Were you not satisfied with the suggestions in that thread.

[anthony.santana.703.j]

No yes, I was satsified just wondering if the tool is sufficient enough. I even got alignment and everything. I know no tool is perfect which is why I asked the question. Either way, I'm currently stuck trying to dedup my reads(this is a Nextflow process error). Actually im doing it for fun, if we rely on tools made by others all the time eventually theres a blackbox or worse, they end up trying to monetize the propietary information. You also limit yourself to the constraints of their pipeline, theres also no harm in building my own.

[GenoMax]

You also limit yourself to the constraints of their pipeline, theres also no harm in building my own.

Absolutely. Consider this though. Unless you are going to write programs to do everything on your own, you are going to be using programs/code from others that is open and freely available for parts of your pipeline. A lot of this code is released under a license that allows personal/academic use without restrictions.

Even the Qiagen pipeline may use open source software under the covers. They may have wrapped it in a web front end for convenience of end-users who want a convenient interface. Generally in cases like that the specifics of the workflow should be available. If a tool requires a license then it may be "black boxed" and/or have proprietary code.

[anthony.santana.703.j]

I undertand, its the whole "Ive only seen as far as i can because ive stood on the shoulder of giants", I just want familiarity with tools and reasoning and i know that Ill be using tools that other make plenty of times, heck the packcages used for bioinformatics are an instance of that. Im not writing my own tools, this is out of genuine curiosity and to achieve something for myself.