Hard to install the pantas

[ld9866]

Dear developer:

We are engaged in the work of pig pangene, and now we want to use pantas to do some analysis of pantranscription, but we find it very difficult to install pantas. I would like to ask if you can provide a docker version, which will be very convenient for the future spread of this software and reduce a lot of trouble in installation.

I also uploaded the problems we encountered, please help us to look at it.
Best day!
Dong

The code we used:
git clone https://github.com/AlgoLab/pantas.git
mamba create -c bioconda -c conda-forge -n pantas python=3.10 biopython gffutils intervaltree bcftools samtools gffread vg snakemake-minimal
conda activate pantas
bash setup.sh
setup (2).log

[ldenti]

Hi, sorry for that. We'll look into simplifying pantas installation: docker seems the way to go (thanks for the suggestion). We'll keep you posted.

Best,

[ldenti]

Hi, we are currently working on the docker image. But in the meantime, we should have fixed the compilation errors. If you want, you can try running again bash setup.sh after pulling the new changes. Otherwise, you can wait for the docker image.

Best,

[ldenti]

Hi, the docker image is ready to use. We updated the README with the instructions needed to build and run it.

Let us know if you encounter any issue.

Best,

[ld9866]

Dear developer:

Thank you for your efforts. I think it will greatly promote the delivery and application of the software. I will test it as soon as possible to see if it can be successful.

Another question is that I have read your algorithm logic and I find that you are very close to the rpvg software (https://github.com/jonassibbesen/rpvg). Can you tell me your optimization and improvement compared to rpvg in what areas?

Best
Dong

[ldenti]

rpvg is a tool for performing (full) transcript quantification. It works on pangenome so it quantifies haplotype-aware transcripts. We developed pantas to directly quantify alternative splicing events (differentially expressed in two conditions) wrt haplotype-aware transcripts. Both approaches uses vg mpmap as spliced aligner since - as far as we know - it's the only spliced aligner which works on pangenome graphs.

To compare with linear reference-based approaches, rpvg is like rsem for pangenomes, pantas is like rmats for pangenomes. So in our opinion, they are two different approaches that tackle two different problems.

Best,

[ld9866]

Thank you for your answer.

I will have some time to test it in the near future, and I will consult you again if I have any questions.

Best
Dong