conda recipe

[kfuku52]

Currently, it's not a high-priority issue, but we would need this upon publication.

[Hego-CCTB]

For later reference:

https://docs.conda.io/projects/conda-build/en/latest/concepts/recipe.html

https://conda.io/projects/conda-build/en/latest/user-guide/tutorials/building-conda-packages.html

https://conda.io/projects/conda-build/en/latest/user-guide/tutorials/build-pkgs.html

[kfuku52]

@Hego-CCTB I will take care of it if you don't have time.

[Hego-CCTB]

This is something I'd like to do myself, but I'd prioritize other issues over this. If that takes too long, though, you can take over any time you want.

[kfuku52]

@Hego-CCTB Have you had a chance to look into this?

[Hego-CCTB]

I have created a conda recipe and tested the installation locally. Here are my steps:

This is the recipe, saved as "meta.yaml" in the working directory.

package:
  name: amalgkit
  version: "0.12.4"

source:
  git_url: https://github.com/kfuku52/amalgkit

build:
  number: 0
  script: "{{ PYTHON }} -m pip install ."

requirements:
  build:
    - python>=3.10
    - pip
  run:
    - python>=3.10
    - numpy>=2.0.0
    - biopython>=1.7
    - pandas>=2.2.2
    - lxml>=5.2.1
    - seqkit>=2.3.1
    - parallel-fastq-dump>=0.6.7
    - fastp>=0.22.0
    - kallisto>=0.48.0
    - r-ggplot2>=3.5.0
    - bioconductor-pcamethods>=1.90.0
    - bioconductor-ruvseq>=1.32.0
    - bioconductor-sva>=3.46.0
    - bioconductor-edger>=3.40.0
    - r-colorspace>=2.1_0
    - r-rcolorbrewer>=1.1_2
    - r-mass>=7.3_57
    - r-nmf>=0.21.0
    - r-dendextend>=1.16.0
    - r-amap>=0.8_19
    - r-pvclust>=2.2_0
    - r-rtsne>=0.16
    - r-patchwork>=1.2.0

about:
  home: https://github.com/kfuku52/amalgkit/
  license: MIT
  description: "AMALGKIT is a toolkit to integrate RNA-seq data from the NCBI SRA database and from private fastq files to generate unbiased cross-species transcript abundance dataset for a large-scale evolutionary gene expression analysis."

A package can then be built by using conda-build . (conda-build may need to be installed with conda install conda-build). An amalgkit.tar.bz2 file is created by the conda-build command, which can be installed from local via

conda install --use-local amalgkit 

I'm running a test set to see if everything works fine, but if it is, all that's left is to make an Anaconda account and upload the package.

[Hego-CCTB]

the kallisto version that got installed using this recipe (0.51) is causing issues during quant. I'll try to figure out the problem, but maybe I need to restrict the recipe to the latest compatible version. All other amalgkit modules work as expected using this recipe.

[kfuku52]

Ideally, AMALGKIT should adjust its output parsing logic based on the installed version of Kallisto, but please assess the feasibility of implementing this functionality.