JOSS Review: Validity of Supported Methods

[gchure]

Hi! I'm reviewing your paper for JOSS and am opening issues as I progress through my checklist. This issue pertains an issue with supported alignment methods.

I admire the work you've put into supporting a broad range of alignment packages/algorithms, but I would appreciate some clarity on why you've chosen some and omitted others.

For example, you have not included support for kallisto which uses pseudoalignment in quantification of RNAseq data. Is there a reason this is not included?

More importantly, I'm concerned by including Tophat as a valid alignment library at all. The corresponding author of Tophat has repeatedly asked people to stop using the tool as there are other, more advanced tools now.

In general, I think there should be some explanation of your process for deciding what is supported by AliNE.

[Juke34]

Thank you for your review and valuable feedbacks.

I have tried to gather the tools I was aware of, and forgot kallisto. I didn't know it was possible to get sam/bam output. Reading the manual I realised my mistake. Thank you for pointing it out. I will try to add it in the next release. My longer-term goal is to gather all possible alignment softwares that provide sam/bam output.

I am aware of the concern with Topcoat but decided to add it with reproducibility in mind. Indeed I wanted to be able to rerun old analyses that used Tophat. With AliNe it will be easy to check and compare the contribution of new tools.

[gchure]

Got, thanks for the clarification. Glad you'll add kallisto -- it's a great tool.

As to Tophat, I understand why you may want to include it for rerunning old analyses, but I still don't think it should be supported at all. As this is open-source software being written to be used by many people with ease, I strongly believe that it shouldn't enable the use of outdated, inefficient methods that are no longer supported. Hisat2 is the successor to tophat, so I really don't see a reason to keep that as an enabled method.

[Juke34]

I have added a warning when Tophat2 is used saying Tophat2 has been deprecated. The developers recommend to switch to HISAT2.

I have also added kallisto

[gchure]

Thanks for adding Kallisto.

A warning does not sufficiently address my concern. This still makes it easy for people to use software that shouldn't be used. Tophat should not be included as a supported alignment method, full stop.