Standard installation through pip and conda #6
Comments
|
I found out that MDAnaysis 2.0 was released at last August. Can you confirm if it contains the features you need and update ProLIF, please? |
|
Hi @lordmiss, Unfortunately no, the base features are there in If you really need a standard installation and you are okay with verifying the output molecule (see the 3 first code cells on this documentation), you could try to clone this repository and change the MDAnalysis requirement in the setup file here to git clone https://github.com/chemosim-lab/ProLIF.git
cd ProLIF
sed -i 's/mdanalysis.*/mdanalysis==2.0.0/' setup.cfg
pip install .Hope this answers your question, |
|
Hi @cbouy, |
|
For the record here is the conda package: conda-forge/staged-recipes#17499 which is currently blocked by mdanalysis (require the 2.1.0 to be released first). |
|
I'm likely to release MDA 2.1.0 over the coming week or so. If this needs to be in 2.1.0 may I ask for someone to push for MDAnalysis/mdanalysis#3044 to finish review? (I've completely lost track of where that PR stands, if it's close to completion then let's aim to get it merged before 2.1.0). |
|
@IAlibay sure, from what I remember it was close to being done, maybe missing a few tests. |
|
Quick update to avoid any confusion: it's now scheduled for MDAnalysis v2.2.0. This way the MDAnalysis core developers can release 2.1.0 very soon without having to wait for me, and 2.2.0 will come out shortly after (technically as soon as I'm done working on the corresponding PR). |
|
Thanks for the update @cbouy. Just ping again once ready, and I will warp up the feedstock PR. |
|
Hi @cbouy Looks like v2.2.0 has just been released! Is there an ETA for the next ProLIF update? |
|
Hi @noahharrison64, yes indeed! |
## [1.0.0] - 2022-06-07 ### Added - Support for multiprocessing, enabled by default (Issue #46). The number of processes can be controlled through `n_jobs` in `fp.run` and `fp.run_from_iterable`. - New interaction: van der Waals contact, based on the sum of vdW radii of two atoms. - Saving/loading the fingerprint object as a pickle with `fp.to_pickle` and `Fingerprint.from_pickle` (Issue #40). ### Changed - Molecule suppliers can now be indexed, reused and can return their length, instead of being single-use generators. ### Fixed - ProLIF can now be installed through pip and conda (Issue #6). - If no interaction is detected in the first frame, `to_dataframe` will not complain about a `KeyError` anymore (Issue #44). - When creating a `plf.Fingerprint`, unknown interactions will no longer fail silently.
|
ProLIF 1.0.0 is now available on both pip and conda-forge! |
TL;DR: don't
pip install prolifyetProLIF currently depends on a custom fork of MDAnalysis which is not available on the PyPI repositories. This is because ProLIF depends on features that are not yet merged in the main branch of MDAnalysis, and that will be available in one of their next release.
Until all the needed features are released on PyPI, I have to rely on this custom fork, which makes it impossible to upload ProLIF v0.3.0 and above to PyPI (see this build for more technical info).
The package that is currently installed with
pip install prolifis a much older version that corresponds to this GitHub repository that I decided to deprecate as it has bugs, is untested, and could only read some MOL2 files.Once the MDAnalysis update is on PyPI, I'll upload the releases there. For
conda, I'm guessing the issue will be the same (I doubt that I can have dependencies that aren't on conda/pip) so I'll wait for this as well.For now, stay safe and
pip install git+https://github.com/chemosim-lab/ProLIF.git, or even better, refer to the installation instructions.The text was updated successfully, but these errors were encountered: