GraphPro is a versatile and pluggable OO python library designed for leveraging deep graph learning representations to gain insights into structural proteins and their conformations
Home Page: https://graphpro.readthedocs.io/en/latest/
License: GNU General Public License v3.0
For members using conda the project should be available to install without using pip.
https://docs.conda.io/projects/conda-build/en/latest/user-guide/tutorials/build-pkgs.html
Spare implementation of pytorch complains about edge index floats, this is truly wasteful for my undertanding.
This implementation requires networkx to transform a graph to a Data() object, remove this requirement to avoid the extra level of indirection.
The project lisence is a GNU license were the medatadata of the project indicates a MIT,
Adjust the metadata as GNU is been use previously by @alepandini, as it covers potential patents.
When a graph is generated it should allow us to translated to PyG
The trnaformation require a set of functions to tranform node / edge attributes to vectors.
GraphCol can be added to a specific dataset but this out of the scope for this ticket.
Ensure to_data can use transformations to encode node attributes.
We can create a specific IMemoryDataset
from torch_geometric.data import InMemoryDataset
from typing import Callable, Optional
class Dataset(InMemoryDataset):
def __init__(
self,
root: str,
collection: GraphCollection,
transform: Optional[Callable] = None,
pre_transform: Optional[Callable] = None,
pre_filter: Optional[Callable] = None,
):
super().__init__(root, transform, pre_transform, pre_filter)
self.data, self.slices = self.collate([g.to_data() for g in collection])
Include instructors and requirmeents for installation
Example: https://pytorch-geometric.readthedocs.io/en/latest/install/installation.html
Given a collection of graphs, that can be exposed to a dataset:
dataset = col.to_dataset('.', node_encoders=[ResidueType()])
Node encoders provide a set of annotation encoders that transform node properties into an X tensor, additionally, we need a property target=Target, where Target can be a node target or a edge target ( link prediction )
When a NodeTarget it performs an encoding of a specific node property into a tensor.
As part of the documentation push, we need to have a read the doc site that contains a basic skelethon
Tasks:
To enable branch projection, we expect actions to validate the PR.
Configure PR branch protection to verify this test.
include a getstarting page in the document
Step/Step representation of graph generation using a basic ContactMap, use demo notebook as example.
Using main branch:
when a trajectory file do not contain occupancy,
and a developer try to generate graph
then graph generation file due expected occupancy information.
src/graphpro/graphgen.py:32: in generate
G = rep.generate(self.ag, self.name)
src/graphpro/graphgen.py:20: in generate
ca_position = ag.c_alphas_positions(self.chain)
src/graphpro/model.py:39: in c_alphas_positions
return self._c_alphas(chain).positions
src/graphpro/model.py:29: in _c_alphas
atoms_filtered = [ag[ag.occupancies.tolist().index(
src/graphpro/model.py:29: in <listcomp>
atoms_filtered = [ag[ag.occupancies.tolist().index(
../ml/lib/python3.10/site-packages/MDAnalysis/core/groups.py:2537: in __getattr__
return super(AtomGroup, self).__getattr__(attr)
_ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _
self = <AtomGroup with 1 atom>, attr = 'occupancies'
def __getattr__(self, attr):
selfcls = type(self).__name__
if attr in _TOPOLOGY_ATTRS:
cls = _TOPOLOGY_ATTRS[attr]
if attr == cls.singular and attr != cls.attrname:
err = ('{selfcls} has no attribute {attr}. '
'Do you mean {plural}?')
raise AttributeError(err.format(selfcls=selfcls, attr=attr,
plural=cls.attrname))
else:
err = 'This Universe does not contain {singular} information'
> raise NoDataError(err.format(singular=cls.singular))
E MDAnalysis.exceptions.NoDataError: This Universe does not contain occupancy information
React
A declarative, efficient, and flexible JavaScript library for building user interfaces.
🖖 Vue.js is a progressive, incrementally-adoptable JavaScript framework for building UI on the web.
Typescript
TypeScript is a superset of JavaScript that compiles to clean JavaScript output.
An Open Source Machine Learning Framework for Everyone
Django
The Web framework for perfectionists with deadlines.
Laravel
A PHP framework for web artisans
Bring data to life with SVG, Canvas and HTML. 📊📈🎉
JavaScript (JS) is a lightweight interpreted programming language with first-class functions.
Some thing interesting about web. New door for the world.
A server is a program made to process requests and deliver data to clients.
Machine learning is a way of modeling and interpreting data that allows a piece of software to respond intelligently.
Some thing interesting about visualization, use data art
Some thing interesting about game, make everyone happy.
Facebook
We are working to build community through open source technology. NB: members must have two-factor auth.
Microsoft
Open source projects and samples from Microsoft.
Google
Google ❤️ Open Source for everyone.
Alibaba
Alibaba Open Source for everyone
D3
Data-Driven Documents codes.
Tencent
China tencent open source team.