# PyDTI -- a Python library for drug-target interaction prediction
version 1.0, June 08 2015
--------
This package is written by:
Yong Liu
Email: liuy0054@e.ntu.edu.sg
School of Computer Engineering, Nanyang Technological University, Singapore
Joint NTU-UBC Research Centre of Excellence in Active Living for Elderly (LILY), Nanyang Technological University, Singapore
For any questions regarding to this library, please feel free to contact the author.
--------
PyDTI is implemented by Python 2.7.9, which can be downloaded from: https://www.python.org/downloads/release/python-279/. PyDTI requires several other Python packages, including Numpy, scikit-learn, Scipy, and pymatbridge (a bridge between Python and Matlab).
The original packages can be found here:
http://www.numpy.org/
http://scikit-learn.org/stable/
http://www.scipy.org/
http://arokem.github.io/python-matlab-bridge/
Note that pymatbridge is only required by KBMF2K. The 64-bit Windows binaries of Numpy, scikit-learn, and Scipy can also be found at: http://www.lfd.uci.edu/~gohlke/pythonlibs/.
1. Add the folder "$PYTHON_ROOT$/Scipts/" to the system path. Please replace "$PYTHON_ROOT$" with the root folder of Python in your system.
2. Install the packages using pip utility. Open a console and type the following to install
pip install numpy scipy scikit-learn
--------
For the KBMF2K method, please download the matlab code implemented by M. Gonen from http://users.ics.aalto.fi/gonen/bioinfo12.php and put the matlab code in the sub-folder "kbmf2k/". To connect the Matlab code and Python, we need to define a Matlab function:
function predictR = kbmf(args)
Kx = args.Kx;
Kz = args.Kz;
Y = args.Y;
R = args.R;
state = kbmf_regression_train(Kx, Kz, Y, R);
prediction = kbmf_regression_test(Kx, Kz, state);
predictR = prediction.Y.mu;
end
Save this function into a Matlab file named kbmf.m and put this file into the subfolder "kbmf2k/".
--------
To get the results of different methods, please run PyDTI.py by setting suitable values for the following parameters:
--method set DTI prediction method
--dataset: choose the benchmark dataset, i.e., nr, gpcr, ic, e
--folder: set the the folder that contains the datasets (default "datasets/")
--csv: choose the cross-validation setting, 1 for CVS1, 2 for CVS2, and 3 for CVS3, (default 1)
--specify-arg: 0 for choosing optimal arguments, 1 for using default/specified arguments (default 1)
--method-opt: set arguments for each method (method ARGUMENTS have the form name=value)
--predict-num: 0 for not predicting novel DTIs, a positive integer for predicting top-N novel DTIs (default 0)
Here are some examples:
(1) run a method with default arguments
python PyDTI.py --method="nrlmf" --dataset="nr"
python PyDTI.py --method="nrlmf" --dataset="nr" --cvs=2
python PyDTI.py --method="nrlmf" --dataset="nr" --cvs=2 --specify-arg=1
(2) run a method with specified arguments
python PyDTI.py --method="nrlmf" --dataset="nr" --cvs=1 --specify-arg=1 --method-opt="r=100"
python PyDTI.py --method="nrlmf" --dataset="nr" --cvs=1 --specify-arg=1 --method-opt="c=5 K1=5 K2=5 r=100 lambda_d=0.125 lambda_t=0.125 alpha=0.25 beta=0.125 theta=0.5"
You can refer to lines 47-58 in the PyDTI.py for the default parameters of each DTI prediction method.
(3) choose the optimal parameters for a method
python PyDTI.py --method="nrlmf" --dataset="nr" --cvs=1 --specify-arg=0
(4) predict the top-100 novel DTIs
python PyDTI.py --method="nrlmf" --dataset="nr" --predict-num=100 --method-opt="r=100"
4. You can run sat_analysis.py for the statistical comparision between NRLMF and other baseline methods on all datasets, under different cross-validation settings. Note that you should first obtain the auc and aupr results of each method.
python sta_analysis.py
没有合适的资源?快使用搜索试试~ 我知道了~
使用wiZAN和其他算法预测化学蛋白质关联附matlab代码.zip
共588个文件
txt:189个
csv:64个
tsv:52个
1.该资源内容由用户上传,如若侵权请联系客服进行举报
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
2.虚拟产品一经售出概不退款(资源遇到问题,请及时私信上传者)
版权申诉
0 下载量 113 浏览量
2024-01-26
21:33:51
上传
评论
收藏 151.98MB ZIP 举报
温馨提示
1.程序语言为matlab,程序可出预测效果图,迭代优化图,相关分析图,运行环境matlab2020b及以上。 2.代码特点:参数化编程、参数可方便更改、代码编程思路清晰、注释明细。 3.适用对象:计算机,电子信息工程、数学等专业的大学生课程设计、期末大作业和毕业设计。 4.作者介绍:某大厂资深算法工程师,从事Matlab算法仿真工作10年;擅长智能优化算法、神经网络预测、信号处理、元胞自动机等多种领域的算法仿真实验,更多仿真源码、数据集定制私信+。
资源推荐
资源详情
资源评论
收起资源包目录
使用wiZAN和其他算法预测化学蛋白质关联附matlab代码.zip (588个子文件)
clustalo-LbGAPTemplates_comp155_c0_seq1.ali 8KB
clustalo-Lbtemp_pdbseq.ali 6KB
comp155_c0_seq1.ali 3KB
clustalo-Lbtemp_pdbseq.pir.bak 3KB
zc_chem_chem_sim05.csv 45.41MB
cyp450_chems_ExtFP.csv 33.66MB
cyp450_chems_PubChemFP.csv 28.99MB
utr_utr.csv 20.48MB
interactions.csv 19.56MB
interactions.csv 19.56MB
CYP450_Assays.csv 13.06MB
CYP450_Assays.csv 13.06MB
cyp450_chems_SubstructFP.csv 10.21MB
zc_active.csv 2.48MB
train5.csv 2.44MB
train3.csv 2.44MB
train4.csv 2.44MB
train6.csv 2.44MB
train2.csv 2.44MB
train10.csv 2.43MB
train9.csv 2.43MB
train8.csv 2.43MB
train1.csv 2.43MB
train7.csv 2.43MB
cyp1a2.csv 1.3MB
cyp2c19.csv 1.29MB
cyp3a4.csv 1.17MB
cyp2c9.csv 920KB
zc_inactive.csv 859KB
cyp2d6.csv 623KB
DC_prot_prot_sim.csv 569KB
CYP450Chembl_chemIdx.csv 459KB
CYP450Chembl_chemIdx.csv 459KB
CYP450_chembl_chem_prot.csv 426KB
ZC_ambiguous_pairs_analysis.csv 336KB
test7.csv 278KB
test1.csv 277KB
test8.csv 277KB
test9.csv 277KB
test10.csv 277KB
test2.csv 277KB
test6.csv 277KB
test4.csv 277KB
test3.csv 277KB
test5.csv 277KB
cyp450_ChemIdx.csv 250KB
CYP450_Profeat.csv 145KB
FDA_approved_drugs_inchikey.csv 47KB
zc_ambiguous.csv 45KB
drug_infor.csv 44KB
prw_chem_prot_test_tc06_wizan.csv 43KB
prw_chem_prot_test_tc1_wizan.csv 41KB
FDA_drugpair_tani0_cosine0.99up.csv 40KB
DC_chem_chem_sim05.csv 39KB
prw_chem_prot_test_tc09_wizan.csv 37KB
prw_chem_prot_test_tc08_wizan.csv 37KB
prw_chem_prot_test_tc07_wizan.csv 37KB
rbp_rbp.csv 28KB
prot_prot_chembl_blast.csv 21KB
DC_chem_prot_wiZAN_predictedScore.csv 3KB
DC_chem_prot_association_index.csv 1KB
human_cyp_features_rnacomm.csv 627B
csvout.csv 431B
cyp_blast_sim.csv 158B
train.csv 64B
train_demo.csv 35B
test.csv 32B
test_demo.csv 10B
comp155_c0_seq1.D00000001 446KB
comp155_c0_seq3.D00000001 445KB
comp155_c0_seq1.D00000001 230KB
comp155_c0_seq1.D00000001 141KB
comp155_c0_seq1.D00000002 448KB
comp155_c0_seq3.D00000002 444KB
comp155_c0_seq1.D00000002 229KB
comp155_c0_seq1.D00000003 449KB
comp155_c0_seq3.D00000003 447KB
comp155_c0_seq1.D00000003 445KB
comp155_c0_seq3.D00000004 445KB
comp155_c0_seq3.D00000005 446KB
comp155_c0_seq1.DL00200001 84KB
comp155_c0_seq1.DL00210001 84KB
comp155_c0_seq1.DL00220001 84KB
comp155_c0_seq1.DL00230001 84KB
comp155_c0_seq1.DL00240001 84KB
comp155_c0_seq1.DL00250001 84KB
glmnetMex.dvi 7KB
GLMnet.f 555KB
glmnetMex.F 38KB
DC_newprot_fasta.fas 98KB
CYP450_Targets.fas 16KB
human_cyp.fas 3KB
human_cyp.fas 3KB
LbGAP_LhGAP_rhogap_and_signalPcontainings.fas 2KB
comp155_c0_seq1_SSP_SYMPRED.fas 630B
comp155_c0_seq1.fas 415B
comp155_c0_seq3_RhoGAP.fas 240B
ChemFeaturesHDF5.tar.gz 5.63MB
ChemFeatureHDF5.tar.gz 5.62MB
SubstructFP.h5 40.42MB
共 588 条
- 1
- 2
- 3
- 4
- 5
- 6
资源评论
matlab科研助手
- 粉丝: 1w+
- 资源: 1734
上传资源 快速赚钱
- 我的内容管理 展开
- 我的资源 快来上传第一个资源
- 我的收益 登录查看自己的收益
- 我的积分 登录查看自己的积分
- 我的C币 登录后查看C币余额
- 我的收藏
- 我的下载
- 下载帮助
最新资源
- 基于STM32F103C8T6单片机蓄电池在线监测系统主板硬件(原理图+PCB)工程文件.zip
- mysql大纲资料.txt
- c++大纲资料.txt
- 效率工具bat脚本实现日志提取
- MyBatis 中动态 SQL 的示例
- STM8L101F3P6单片机+CC1100模块433M遥控器设计硬件(原理图+PCB)工程文件.zip
- 上传下载铁人下载系统 Liuxing 1.0-liuxing1.0.rar
- 南京邮电大学数学实验实力雄厚,凭借其优秀的师资力量、丰富的实践教学资源和卓越的科研成果,成为国内一流的数学实验教学和科研基地
- 【火爆朋友圈的今天吃什么源码 v1.0】随机的为用户带来每一天的用餐选择和推荐.rar
- MPU6050中文版数据手册
资源上传下载、课程学习等过程中有任何疑问或建议,欢迎提出宝贵意见哦~我们会及时处理!
点击此处反馈
安全验证
文档复制为VIP权益,开通VIP直接复制
信息提交成功