You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
PaddleHelix is a machine-learning-based bio-computing framework aiming at facilitating the development of the following areas:
13
-
> * Vaccine design
14
-
> * Drug discovery
15
-
> * Precision medicine
12
+
## Latest News
13
+
`2021.06.17` PaddleHelix team won the 2nd place in the [OGB-LCS KDD Cup 2021 PCQM4M-LSC track](https://ogb.stanford.edu/kddcup2021/results/), predicting DFT-calculated HOMO-LUMO energy gap of molecules. Please refer to [the solution](./competition/kddcup2021-PCQM4M-LSC) for more details.
16
14
17
-
## Features
18
-
* High Efficiency: We provide LinearRNA, a highly efficient toolkit for RNA structure prediction and analysis. LinearFold & LinearPartition achieve O(n) complexity in RNA-folding prediction, which is hundreds of times faster than traditional folding techniques.
* Large-scale Representation Learning: Self-supervised learning for molecule representations offers prospects of a breakthrough in tasks with limited annotation, including drug profiling, drug-target interaction, protein-protein interaction, RNA-RNA interaction, protein folding, RNA folding, and molecule design. PaddleHelix implements various representation learning algorithms and state-of-the-art large-scale pre-trained models to help developers start from "the shoulders of giants" quickly.
`2021.05.20` PaddleHelix v1.0 released. 1) Update from static framework to dynamic framework; 2) Add new applications: molecular generation and drug-drug synergy.
27
16
28
-
* Rich examples and applications: PaddleHelix provides frequently used components such as networks, datasets, and pre-trained models. Users can easily use those components to build up their models and systems. PaddleHelix also provides multiple applications, such as compound property prediction, drug-target interaction, and so on.
17
+
`2021.03.15`PaddleHelix team rank 1st in the ogbg-molhiv and ogbg-molpcba of [OGB](https://ogb.stanford.edu/docs/leader_graphprop/), predicting the molecular properties.
29
18
30
-
----
19
+
---
31
20
32
-
## Installation
21
+
## Introduction
22
+
PaddleHelix is a bio-computing tools, taking advantage of machine learning approach, especially deep neural networks, for facilitating the development of the following areas:
23
+
***Drug Discovery**. Provide 1) Large-scale pre-training models: compounds and proteins; 2) Various applications: molecular property prediction, drug-target affinity prediction, and molecular generation.
24
+
***Vaccine Design**. Provide RNA design algorithms, including LinearFold and LinearPartition.
25
+
***Precision Medicine**. Provide application of drug-drug synergy.
33
26
34
-
The installation prerequisites and guide can be found [here](./installation_guide.md).
[PaddleHelix platform](https://paddlehelix.baidu.com/) provides the AI + biochemistry abilities for the scenarios of drug discovery, vaccine design and precision medicine.
37
34
38
-
## Documentation
35
+
### Installation Guide
36
+
PaddleHelix is a bio-computing repository based on [PaddlePaddle](https://github.com/paddlepaddle/paddle), a high-performance Parallelized Deep Learning Platform. The installation prerequisites and guide can be found [here](./installation_guide.md).
39
37
40
38
### Tutorials
41
-
* We provide abundant [tutorials](./tutorials) to help you navigate the repository and start quickly.
42
-
* PaddleHelix is based on [PaddlePaddle](https://github.com/paddlepaddle/paddle), a high-performance Parallelized Deep Learning Platform.
39
+
We provide abundant [tutorials](./tutorials) to help you navigate the repository and start quickly.
40
+
***Drug Discovery**
41
+
-[Compound Representation Learning and Property Prediction](./tutorials/compound_property_prediction_tutorial.ipynb)
42
+
-[Protein Representation Learning and Property Prediction](./tutorials/protein_pretrain_and_property_prediction_tutorial.ipynb)
PaddleHelix team participated in multiple competitions related to bio-computing. The solutions can be found [here](./competition).
54
62
55
63
### Guide for developers
56
-
* If you need help in modifying the source code of PaddleHelix, please see our [Guide for developers](./developer_guide.md).
64
+
* To develope new functions based on the source code of PaddleHelix, please refer to [guide for developers](./developer_guide.md).
65
+
* For more details of the APIs, please refer to the [documents](https://paddlehelix.readthedocs.io/en/dev/).
57
66
58
67
### Welcome to join us
59
68
We are looking for machine learning researchers / engineers or bioinformatics / computational chemistry researchers interested in AI-driven drug design.
PaddleHelix team participated in multiple competitions related to bio-computing.
2
+
3
+
-`2021.06.17` We won the 2nd place in the [OGB-LCS KDD Cup 2021 PCQM4M-LSC track](https://ogb.stanford.edu/kddcup2021/results/), predicting DFT-calculated HOMO-LUMO energy gap of molecules. Please refer to [the solution](./kddcup2021-PCQM4M-LSC) for more details.
4
+
5
+
-`2021.03.15` We 1st in the ogbg-molhiv and ogbg-molpcba of [OGB](https://ogb.stanford.edu/docs/leader_graphprop/), predicting the molecular properties. Please refer to [the solution](./ogbg_molhiv) for more details.
Copy file name to clipboardExpand all lines: docs/readme.rst
-21Lines changed: 0 additions & 21 deletions
Original file line number
Diff line number
Diff line change
@@ -17,27 +17,6 @@ Welcome to PaddleHelix Helper
17
17
:alt:Documentation Status
18
18
19
19
20
-
PaddleHelix is a machine-learning-based bio-computing framework aiming at facilitating the development of the following areas:
21
-
* Vaccine design
22
-
* Drug discovery
23
-
* Precision medicine
24
-
25
-
Features
26
-
========
27
-
28
-
- High Efficiency: We provide LinearRNA, a highly efficient toolkit for RNA structure prediction and analysis. LinearFold & LinearPartition achieve O(n) complexity in RNA-folding prediction, which is hundreds of times faster than traditional folding techniques.
29
-
30
-
.. image:: ../.github/LinearRNA.jpg
31
-
:align:center
32
-
33
-
- Large-scale Representation Learning: Self-supervised learning for molecule representations offers prospects of a breakthrough in tasks with limited annotation, including drug profiling, drug-target interaction, protein-protein interaction, RNA-RNA interaction, protein folding, RNA folding, and molecule design. PaddleHelix implements various representation learning algorithms and state-of-the-art large-scale pre-trained models to help developers start from "the shoulders of giants" quickly.
34
-
35
-
.. image:: ../.github/paddlehelix_features.jpg
36
-
:align:center
37
-
38
-
- Rich examples and applications: PaddleHelix provides frequently used components such as networks, datasets, and pre-trained models. Users can easily use those components to build up their models and systems. PaddleHelix also provides multiple applications, such as compound property prediction, drug-target interaction, and so on.
0 commit comments