空间站广场

论文

Notebooks

比赛

课程

Apps

我的主页

我的Notebooks

我的论文库

我的足迹

我的工作空间

任务

节点

文件

数据集

镜像

项目

数据库

公开

Molecular Vector Representation Based on Uni-Mol

Uni-Mol

Deep Learning

Uni-MolDeep Learning

Yani Guan

更新于 2024-10-24

推荐镜像 :Uni-Mol:unimol-qsar:v0.5

推荐机型 :c3_m4_1 * NVIDIA T4

Molecular Vector Representation Based on Uni-Mol

What is Molecular/Atomic Vector Representation

import Uni-Mol

Molecular Vector Representation Based on Uni-Mol

Open source code: https://github.com/deepmodeling/Uni-Mol
Paper link: https://openreview.net/forum?id=6K2RM6wVqKu

What is Molecular/Atomic Vector Representation

Molecular/atomic level vector representation is the process of representing the chemical and physical properties of atoms or molecules as mathematical vectors. This is significant in cheminformatics, computational chemistry, drug design, and materials science.
Atomic level vector representation includes information such as atom type, electronegativity, atomic radius, electron shell structure, charge distribution, and bond type. For example, a carbon atom's vector might be [6, 2.55, 0.77, [2, 4], 0.0, 1].
Molecular level vector representation covers molecular structure information, electronic properties, geometric structure, spectral properties, thermodynamic properties, and reactivity. For example, a water molecule's vector might be [18.015, 1.85, [0.957, 104.5], [0.34, 0.17, 1.85], -75.0].

This representation method is widely used in machine learning models, molecular similarity searches, and chemical reaction predictions. It provides an important foundation for quantitative analysis and computation, promoting research and applications in chemistry and biology.

代码

文本

import Uni-Mol

代码

文本

[1]

# Select the image: unimol-qsar:unimollastest, choose the GPU machine type

# Import unimol

from unimol import UniMolRepr

import numpy as np

import pandas as pd

/opt/conda/lib/python3.8/site-packages/tqdm/auto.py:22: TqdmWarning: IProgress not found. Please update jupyter and ipywidgets. See https://ipywidgets.readthedocs.io/en/stable/user_install.html
  from .autonotebook import tqdm as notebook_tqdm

代码

文本

[2]

import numpy as np

from unimol_tools import UniMolRepr

# single smiles unimol representation

clf = UniMolRepr(data_type='molecule', remove_hs=False)

smiles = 'c1ccc(cc1)C2=NCC(=O)Nc3c2cc(cc3)[N+](=O)[O]'

smiles_list = [smiles]

unimol_repr = clf.get_repr(smiles_list, return_atomic_reprs=True)

# CLS token repr

print(np.array(unimol_repr['cls_repr']).shape)

# atomic level repr, align with rdkit mol.GetAtoms()

print(np.array(unimol_repr['atomic_reprs']).shape)

---------------------------------------------------------------------------
ModuleNotFoundError                       Traceback (most recent call last)
Input In [2], in <cell line: 2>()
      1 import numpy as np
----> 2 from unimol_tools import UniMolRepr
      3 # single smiles unimol representation
      4 clf = UniMolRepr(data_type='molecule', remove_hs=False)

ModuleNotFoundError: No module named 'unimol_tools'

代码

文本

[ ]

# Single SMILES UniMol representation

clf = UniMolRepr(data_type='molecule', remove_hs=False)

smiles = 'c1ccc(cc1)C2=NCC(=O)Nc3c2cc(cc3)[N+](=O)[O]'

smiles_list = [smiles]

unimol_repr = clf.get_repr(smiles_list, return_atomic_reprs=True)

# Uni-Mol molecular representation using the cls token

print(np.array(unimol_repr['cls_repr']).shape)

# Uni-Mol atomic-level representation, consistent with the atom order in rdkit Mol

print(np.array(unimol_repr['atomic_reprs']).shape)

代码

文本

[ ]

%%bash

# download sample dataset, CNS drug data

rm -rf mol_train.csv

wget -nv https://bohrium-example.oss-cn-zhangjiakou.aliyuncs.com/unimol-qsar/mol_train.csv

代码

文本

[ ]

smiles_list = pd.read_csv('mol_train.csv')['SMILES'].to_list()

y = pd.read_csv('mol_train.csv')['TARGET'].to_list()

repr_dict = clf.get_repr(smiles_list)

unimol_repr_list = np.array(repr_dict['cls_repr'])

代码

文本

[ ]

print(unimol_repr_list.shape)

代码

文本

[ ]

import numpy as np

import matplotlib.pyplot as plt

from mpl_toolkits.mplot3d import Axes3D

from sklearn.decomposition import PCA

代码

文本

[ ]

pca = PCA(n_components=2)

X_reduced = pca.fit_transform(unimol_repr_list)

代码

文本

[ ]

# visualization

colors = ['r', 'g', 'b']

markers = ['s', 'o', 'D']

labels = ['Target:0','Target:1']

plt.figure(figsize=(8, 6))

for label, color, marker in zip(np.unique(y), colors, markers):

plt.scatter(X_reduced[y == label, 0],

X_reduced[y == label, 1],

c=color,

marker=marker,

label=labels[label],

edgecolors='black')

plt.xlabel('Principal Component 1')

plt.ylabel('Principal Component 2')

plt.legend(loc='best')

plt.title('Unimol Repr')

plt.show()

代码

文本

[ ]

代码

文本

Uni-Mol

Deep Learning

Uni-MolDeep Learning

点个赞吧