artic-network / fieldbioinformatics

The ARTIC field bioinformatics pipeline
MIT License
110 stars 69 forks source link

Conda environment set up #130

Open zkkhalil opened 9 months ago

zkkhalil commented 9 months ago

Hello,

I am attempting to set up the conda environment and I keep running into incompatibility issues

Output: warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY warning libmamba Problem type not implemented SOLVER_RULE_STRICT_REPO_PRIORITY Could not solve for environment specs The following packages are incompatible ├─ artic-tools 0.2.6** is installable and it requires │ └─ boost-cpp >=1.70.0,<1.70.1.0a0 with the potential options │ ├─ boost-cpp 1.70.0 would require │ │ └─ icu >=67.1,<68.0a0 , which can be installed; │ ├─ boost-cpp 1.70.0 would require │ │ └─ icu >=64.2,<65.0a0 , which can be installed; │ └─ boost-cpp 1.70.0 would require │ └─ icu >=58.2,<59.0a0 , which can be installed; └─ medaka 1.8.0** is not installable because it requires └─ tensorflow >=2.8.0,<2.9.dev0 but there are no viable options ├─ tensorflow [2.8.0|2.8.1] would require │ └─ tensorflow-base [2.8.0 cpu_py310h75e90da_0|2.8.0 cpu_py37h4373017_0|...|2.8.1 cpu_py39h45807a0_0], which requires │ └─ icu >=70.1,<71.0a0 , which conflicts with any installable versions previously reported; ├─ tensorflow [2.8.0|2.8.1] would require │ └─ __cuda, which is missing on the system; └─ tensorflow 2.8.2 conflicts with any installable versions previously reported.

I tried to play around with the versions using python 3.8.0 and not specifying a certain version of medaka (leaving it as medaka. However, I then ran into the following problem:

File "/sc/arion/work/khaliz03/snakemake/c6c19c46c4dc4f596c4635e3613cc2a1_/lib/python3.8/site-packages/h5py/__init__.py", line 46, in <module> from ._conv import register_converters as _register_converters File "h5py/_conv.pyx", line 1, in init h5py._conv File "h5py/h5t.pyx", line 293, in init h5py.h5t File "/sc/arion/work/khaliz03/snakemake/c6c19c46c4dc4f596c4635e3613cc2a1_/lib/python3.8/site-packages/numpy/__init__.py", line 320, in __getattr__ raise AttributeError("module {!r} has no attribute " AttributeError: module 'numpy' has no attribute 'typeDict'

BioWilko commented 9 months ago

This very much looks like a problem with your conda installation, what hardware are you trying to run this snakemake pipeline on?

There are existing nextflow pipelines e.g. https://github.com/connor-lab/ncov2019-artic-nf or https://github.com/epi2me-labs/wf-artic which run fieldbioinformatics for you which may do what you need.

zkkhalil commented 9 months ago

It is an internal pipeline that is being run on an HPC cluster. We previously used the YAML environment set up in the artic-ncov2019 repository because we use it for targeted reference-based assembly of SARS-CoV-2. However, due to the new medaka versions, we kept running into trouble when we had to set up the environment again. The internal pipeline that uses snakemake sets up all necessary environments using their respective YAML files. The only issue that arises is with the artic set up.

BioWilko commented 9 months ago

Can you point me to the repository for this pipeline? I'm specifically interested in the yaml environment definition.