中醫藥在亞洲許多國家具有數千年歷史，近年來甚至逐漸受到歐美國家的重視，但相對於西方醫學，中醫藥在現代醫學或藥理研究仍較為缺乏，其原因除傳統中醫藥典籍多為古文，還有中西醫語言與系統上的歧異，無法將西醫概念直接推展到中藥的系統中，而隨著中醫藥科學上的發展與研究，許多新療法或新藥也被發掘，中醫藥的科學化勢必為未來趨勢，且極具醫學潛力。此研究為預測中藥適應症，嘗試以西藥資料拓展至中藥概念，以實踐多成分、多標靶的藥物模型，適應症為藥物治療特定疾病或醫療狀況，相對於疾病概念更為廣泛，過去相關研究多為預測藥物疾病，但對於中藥而言，適應症廣泛的概念更為合適。為了彌補中藥適應症資料的缺乏，我們嘗試了兩種文字探勘的方法以挖掘中藥適應症，以及手動標註中藥方劑適應症。為了預測藥物適應症，我們建構了包含藥物、標靶路徑與適應症的多元網絡，以隨機漫步演算法進行預測。對於西藥適應症預測，以十次交叉驗證，AUC為0.948±0.019，優於過去研究，對於中藥方劑適應症預測，AUC為0.745。僅管中藥適應症預測結果並不佳，但模型可應用於中藥或植物藥有效成分的篩選，以綠茶茶多酚抗癌效果預測為例，結果顯示的確可以篩選出最佳抗癌成分(EGCG)。

In Asia, traditional Chinese medicine (TCM) have been evolved over thousand years, and becoming attractive to Western in decade. However, the evidence and research works of its effectiveness are still limited by its high complexity and difficulties in analysis.The purpose of our research was to predict TCM-indication based on modern medicine networks that was constructed for multi-ingredients and multi-targets.
Meaning of indication is a disease or specific medical condition for medication. Although much research of drug prediction focused on disease, indication is comprehensive and appropriate for TCM prediction. In order to make up for the lack of TCM-indication associations, we used two types of text mining to discover herb indication and created the label of TCM prescription indication by hand.
To predict indication, we used random walk model based on a multi-network containing drug, target/pathway and indication. Our indication prediction model for single drug achieved 0.948±0.019 AUC in 10-fold cross-validation, its performance is better than previous research. Our model indication prediction for TCM prescription achieved 0.745 AUC, its performance is not good enough to work in clinical medicine.
However, our model can be applied to discover candidate drugs from herb. As a case in point, anticancer prediction for polyphenols green tea. The result showed that the model can select the most effective compound, epigallocatechin eallate (EGCG).

摘要 i
Abstract ii
List of Tables vi
List of Figures vii
1 研究目的 1
2 文獻探討 3
2.1 藥物適應症(Indications for Drugs) . . . . . . . . . . . . . . . . . . . . . 3
2.2 中藥生物特性與發展 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3
2.3 藥物相關網絡建構. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 4
2.3.1 基於化學(chemical-based) . . . . . . . . . . . . . . . . . . . . . . 5
2.3.2 基於蛋白質標靶與生物路徑(target-based and pathway-based) . . 5
2.3.3 基於適應症 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 6
3 研究方法 7
3.1 資料集 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 8
3.1.1 黃金標準資料(gold standard data)處理 . . . . . . . . . . . . . . . 9
3.1.2 中藥測試資料 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2 建構網路 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10
3.2.1 藥物對藥物同質網路 . . . . . . . . . . . . . . . . . . . . . . . . 11
3.2.2 標靶對標靶同質網路 . . . . . . . . . . . . . . . . . . . . . . . . 12
3.2.3 路徑對路徑同質網路 . . . . . . . . . . . . . . . . . . . . . . . . 13
3.2.4 適應症對適應症同質網路 . . . . . . . . . . . . . . . . . . . . . 15
3.2.5 藥物對標把異質網路 . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2.6 藥物對路徑異質網路 . . . . . . . . . . . . . . . . . . . . . . . . 15
3.2.7 藥物對適應症異質網路. . . . . . . . . . . . . . . . . . . . . . . 16
3.2.8 標靶對適應症異質網路. . . . . . . . . . . . . . . . . . . . . . . 16
3.2.9 路徑對適應症異質網路. . . . . . . . . . . . . . . . . . . . . . . 16
3.3 多元網路隨機漫步(Random walk on multiplex networks) . . . . . . . . . 17
3.4 優化隨機漫步 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 19
3.4.1 投影矩陣(projected matrix) . . . . . . . . . . . . . . . . . . . . . 19
3.4.2 跳躍參數(jump parameter) . . . . . . . . . . . . . . . . . . . . . . 19
3.5 處理中藥適應症資料 . . . . . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5.1 中藥文獻提取適應症 . . . . . . . . . . . . . . . . . . . . . . . . 20
3.5.2 PubMed文獻挖掘中藥適應症. . . . . . . . . . . . . . . . . . . . 22
3.5.3 中藥方劑文獻處理 . . . . . . . . . . . . . . . . . . . . . . . . . 23
4 實驗設計與結果 26
4.1 以黃金標準資料(gold standard data)驗證模型 . . . . . . . . . . . . . . . 26
4.1.1 輸入測試資料(testing data set) . . . . . . . . . . . . . . . . . . . 26
4.1.2 參數調整. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.1.3 結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27
4.2 以中藥與中藥方劑預測中藥適應症 . . . . . . . . . . . . . . . . . . . . 30
4.2.1 輸入測試資料(testing data set) . . . . . . . . . . . . . . . . . . . 30
4.2.2 結果 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 31
4.3 篩選潛在植物藥有效成分. . . . . . . . . . . . . . . . . . . . . . . . . . 33
4.3.1 輸入預測化合物. . . . . . . . . . . . . . . . . . . . . . . . . . . 34
4.3.2 結果比較. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 35
5 討論與結論 37
5.1 黃金標準資料(gold standard data)驗證模型 . . . . . . . . . . . . . . . . 37
5.2 預測中藥適應症 . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
5.3 未來可能研究方向. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38
References 40
.1 Appendix:方劑處方內容. . . . . . . . . . . . . . . . . . . . . . . . . . . 51
.2 Appendix:方劑標記適應症 . . . . . . . . . . . . . . . . . . . . . . . . . 58

[1] Fei Wang, Ping Zhang, Nan Cao, Jianying Hu, and Robert Sorrentino. Explor-ing the associations between drug side-effects and therapeutic indications.Journalof Biomedical Informatics, 51:15–23, oct 2014. ISSN 1532-0464. doi: 10.1016/J.JBI.2014.03.014.URL https://www.sciencedirect.com/science/article/pii/S1532046414000811.
[2] Matthew T. Villaume, Eran Sella, Garrett Saul, Robert M. Borzilleri, Joseph Fargnoli,Kathy A. Johnston, Haiying Zhang, Mark P. Fereshteh, T. G. Murali Dhar, and Phil S.Baran. Antroquinonol A: Scalable Synthesis and Preclinical Biology of a Phase 2Drug Candidate.ACS Central Science, 2(1):27–31, jan 2016. ISSN 2374-7943. doi:10.1021/acscentsci.5b00345. URL http://pubs.acs.org/doi/10.1021/acscentsci.5b00345.
[3] Yuh-Mou Sue. The effect of lipocol forte capsules on the pharmacokinetics of nifedip-ine after administering single-dose combination to healthy subjects, 2011. URL https://clinicaltrials.gov/ct2/show/NCT01346657.
[4] 李飛.方劑學.人民衛生出版社, 2011.
[5] J. Zhao, P. Jiang, and W. Zhang. Molecular networks for the study of TCM Phar-macology.Briefings in Bioinformatics, 11(4):417–430, jul 2010.ISSN 1467-5463. doi: 10.1093/bib/bbp063. URL https://academic.oup.com/bib/article-lookup/doi/10.1093/bib/bbp063.
[6] J. P. Tollenaere. The role of structure-based ligand design and molecular modelling indrug discovery.Pharmacy World and Science, 18(2):56–62, 1996. ISSN 0928-1231.doi: 10.1007/BF00579706. URL http://link.springer.com/10.1007/BF00579706.
[7] G. Vriend. WHAT IF: A molecular modeling and drug design program.Jour-nal of Molecular Graphics, 8(1):52–56, mar 1990. ISSN 0263-7855. doi: 10.1016/0263-7855(90)80070-V.URL https://www.sciencedirect.com/science/article/pii/026378559080070V.
[8] S Ekins, J Mestres, and B Testa.In silico pharmacology for drug discovery:methods for virtual ligand screening and profiling.British journal of pharma-cology, 152(1):9–20, sep 2007. ISSN 0007-1188. doi: 10.1038/sj.bjp.0707305.URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC1978274.
[9] A Srinivas Reddy and Shuxing Zhang. Polypharmacology: drug discovery for thefuture.Expert review of clinical pharmacology, 6(1):41–7, jan 2013. ISSN 1751-2441. doi: 10.1586/ecp.12.74. UR Lhttp://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3809828.41
[10] Muhammed A Yldrm, Kwang-Il Goh, Michael E Cusick, Albert-L ́aszl ́o Barab ́asi,and Marc Vidal. Drug—target network.Nature Biotechnology, 25(10):1119–1126,oct 2007. ISSN 1087-0156. doi: 10.1038/nbt1338. URL http://www.nature.com/articles/nbt1338.
[11] David Weininger. SMILES, a Chemical Language and Information System. 1. Intro-duction to Methodology and Encoding Rules.J. Chem. Inf. Comput. Sci. Proc. Ed-inburgh Math. Soc. J. Comput. Chem. Z. Naturforsch. Bull. Soc. Chim. MATCH Rev.Res. Fac. Sci.-Univ. Novi Sad, Math. Ser. Computer Enumeration of Substituted Poly-hexes Comput. Chem, 281413(2715):31–36, 1988. URL https://pubs.acs.org/doi/pdf/10.1021/ci00057a005.
[12] Adri`a Cereto-Massagu ́e, Mar ́ıa Jos ́e Ojeda, Cristina Valls, Miquel Mulero, San-tiago Garcia-Vallv ́e, and Gerard Pujadas. Molecular fingerprint similarity searchin virtual screening.Methods, 71:58–63, jan 2015.ISSN 1046-2023.doi:10.1016/J.YMETH.2014.08.005. URL https://www.sciencedirect.com/science/article/pii/S1046202314002631.
[13] Hanna Eckert and J ̈urgen Bajorath. Molecular similarity analysis in virtual screen-ing: foundations, limitations and novel approaches.Drug Discovery Today, 12(5-6):225–233, mar 2007.ISSN 1359-6446.doi: 10.1016/J.DRUDIS.2007.01.011. URL https://www.sciencedirect.com/science/article/pii/S1359644607000529.
[14] Carlie A. LaLone, Daniel L. Villeneuve, Lyle D. Burgoon, Christine L. Russom,Henry W. Helgen, Jason P. Berninger, Joseph E. Tietge, Megan N. Severson, Jenna E.Cavallin, and Gerald T. Ankley. Molecular target sequence similarity as a basis forspecies extrapolation to assess the ecological risk of chemicals with known modes ofaction.Aquatic Toxicology, 144-145:141–154, nov 2013. ISSN 0166-445X. doi: 10.1016/J.AQUATOX.2013.09.004. URL https://www.sciencedirect.com/science/article/pii/S0166445X13002312.
[15] Michael J. Keiser, Vincent Setola, John J. Irwin, Christian Laggner, Atheir I. Ab-bas, Sandra J. Hufeisen, Niels H. Jensen, Michael B. Kuijer, Roberto C. Matos,Thuy B. Tran, Ryan Whaley, Richard A. Glennon, J ́erˆome Hert, Kelan L. H.Thomas, Douglas D. Edwards, Brian K. Shoichet, and Bryan L. Roth. Predictingnew molecular targets for known drugs.Nature, 462(7270):175–181, nov 2009.ISSN 0028-0836. doi: 10.1038/nature08506. URL http://www.nature.com/articles/nature08506.
[16] William R Pearson. An introduction to sequence similarity (”homology”) searching.Current protocols in bioinformatics, Chapter 3:Unit3.1, jun 2013. ISSN 1934-340X.doi: 10.1002/0471250953.bi0301s42.URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3820096.
[17] Jiao Li and Zhiyong Lu. Pathway-based drug repositioning using causal inference.BMC bioinformatics, 14 Suppl 1(Suppl 16):S3, 2013. ISSN 1471-2105. doi: 10.1186/1471-2105-14-S16-S3. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3853312.
[18] Silpa Suthram, Joel T. Dudley, Annie P. Chiang, Rong Chen, Trevor J. Hastie,and Atul J. Butte.Network-Based Elucidation of Human Disease Similari-ties Reveals Common Functional Modules Enriched for Pluripotent Drug Tar-gets.PLoS Computational Biology, 6(2):e1000662, feb 2010. ISSN 1553-7358.doi: 10.1371/journal.pcbi.1000662. URL http://dx.plos.org/10.1371/journal.pcbi.1000662.
[19] X. Wang, N. Gulbahce, and H. Yu. Network-based methods for human disease geneprediction.Briefings in Functional Genomics, 10(5):280–293, sep 2011. ISSN 2041-2649. doi: 10.1093/bfgp/elr024. URL https://academic.oup.com/bfg/article-lookup/doi/10.1093/bfgp/elr024.
[20] Ritu Khare, Jiao Li, and Zhiyong Lu. LabeledIn: Cataloging labeled indications forhuman drugs.Journal of Biomedical Informatics, 52:448–456, dec 2014. ISSN 1532-0464. doi: 10.1016/J.JBI.2014.08.004. URL https://www.sciencedirect.com/science/article/pii/S1532046414001853?via{\%}3Dihub.
[21] Chih-Hsuan Wei, Hung-Yu Kao, and Zhiyong Lu.PubTator:a web-basedtext mining tool for assisting biocuration.Nucleic acids research, 41(WebServer issue):W518–22, jul 2013.ISSN 1362-4962.doi:10.1093/nar/gkt441. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3692066.
[22] Allison B McCoy, Adam Wright, Archana Laxmisan, Madelene J Ottosen, Ja-cob A McCoy, David Butten, and Dean F Sittig. Development and evaluation of acrowdsourcing methodology for knowledge base construction: identifying relation-ships between clinical problems and medications.Journal of the American Med-ical Informatics Association, 19(5):713–718, sep 2012.ISSN 1067-5027.doi:10.1136/amiajnl-2012-000852. URL https://academic.oup.com/jamia/article-lookup/doi/10.1136/amiajnl-2012-000852.
[23] Assaf Gottlieb, Gideon Y Stein, Eytan Ruppin, and Roded Sharan. PREDICT:a method for inferring novel drug indications with application to personalizedmedicine.Molecular systems biology, 7:496, jun 2011.ISSN 1744-4292.doi: 10.1038/msb.2011.26. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3159979.
[24] D.S. Wishart, Craig Knox, An Chi Guo, Savita Shrivastava, Murtaza Hassanali,Paul Stothard, Zhan Chang, and Jennifer Woolsey.DrugBank: a comprehen-sive resource for in silico drug discovery and exploration.Nucleic Acids Re-search, 34(90001):D668–D672, jan 2006.ISSN 0305-1048.doi: 10.1093/nar/gkj067. URL https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkj067.
[25] L. M. Schriml, C. Arze, S. Nadendla, Y.-W. W. Chang, M. Mazaitis, V. Felix,G. Feng, and W. A. Kibbe. Disease Ontology: a backbone for disease semanticintegration.Nucleic Acids Research, 40(D1):D940–D946, jan 2012. ISSN 0305-1048. doi: 10.1093/nar/gkr972. URL https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkr972.45
[26] R. Apweiler, Amos Bairoch, Cathy H. Wu, Winona C. Barker, Brigitte Boeck-mann, Serenella Ferro, Elisabeth Gasteiger, Hongzhan Huang, Rodrigo Lopez,Michele Magrane, Maria J. Martin, Darren A. Natale, Claire O’Donovan, NicoleRedaschi, and LaiSu L. Yeh.UniProt:the Universal Protein knowledge-base.Nucleic Acids Research, 32(90001):115D–119, jan 2004.doi: 10.1093/nar/gkh131. URL https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkh131.
[27] Y. Wang, J. Xiao, T. O. Suzek, J. Zhang, J. Wang, and S. H. Bryant. PubChem: a pub-lic information system for analyzing bioactivities of small molecules.Nucleic AcidsResearch, 37(Web Server):W623–W633, jul 2009. ISSN 0305-1048. doi: 10.1093/nar/gkp456. URL https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/gkp456.
[28] M. Kanehisa and Susumu Goto. KEGG: Kyoto Encyclopedia of Genes and Genomes.Nucleic Acids Research, 28(1):27–30, jan 2000. ISSN 13624962. doi: 10.1093/nar/28.1.27. URL https://academic.oup.com/nar/article-lookup/doi/10.1093/nar/28.1.27.
[29] Ruichao Xue,Zhao Fang, Meixia Zhang, Zhenghui Yi, Chengping Wen,and Tieliu Shi.TCMID:traditional Chinese medicine integrative databasefor herb molecular mechanism analysis.Nucleic Acids Research, 41(D1):D1089–D1095, nov 2012. ISSN 0305-1048. doi: 10.1093/nar/gks1100. URL http://academic.oup.com/nar/article/41/D1/D1089/1057998/TCMID-traditional-Chinese-medicine-integrative.46
[30] Jiang Li, Binsheng Gong, Xi Chen, Tao Liu, Chao Wu, Fan Zhang, Chunquan Li,Xiang Li, Shaoqi Rao, and Xia Li. DOSim: an R package for similarity between dis-eases based on Disease Ontology.BMC bioinformatics, 12:266, jun 2011. ISSN 1471-2105. doi: 10.1186/1471-2105-12-266. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3150296.
[31] Haiying Wang, Francisco Azuaje, Olivier Bodenreider, and Joaqu ́ın Dopazo. GeneExpression Correlation and Gene Ontology-Based Similarity: An Assessment ofQuantitative Relationships.IEEE Symposium on Computational Intelligence inBioinformatics and Computational Biology, 2004:25–31, oct 2004.doi: 10.1109/CIBCB.2004.1393927. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC4317290.
[32] Edward A Codling, Michael J Plank, and Simon Benhamou. Random walk modelsin biology.Journal of the Royal Society, Interface, 5(25):813–34, aug 2008. ISSN1742-5689. doi: 10.1098/rsif.2008.0014. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC2504494.
[33] Wei Liu, Chunquan Li, Yanjun Xu, Haixiu Yang, Qianlan Yao, Junwei Han,Desi Shang, Chunlong Zhang, Fei Su, Xiaoxi Li, Yun Xiao, Fan Zhang, MengDai, and Xia Li.Topologically inferring risk-active pathways toward precisecancer classification by directed random walk.Bioinformatics, 29(17):2169–2177, sep 2013.ISSN 1460-2059.doi: 10.1093/bioinformatics/btt373.URL https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btt373.47
[34] Jie Sun, Hongbo Shi, Zhenzhen Wang, Changjian Zhang, Lin Liu, Letian Wang, Wei-wei He, Dapeng Hao, Shulin Liu, and Meng Zhou. Inferring novel lncRNA–diseaseassociations based on a random walk model of a lncRNA functional similarity net-work.Mol. BioSyst., 10(8):2074–2081, jul 2014. ISSN 1742-206X. doi: 10.1039/C3MB70608G. URL http://xlink.rsc.org/?DOI=C3MB70608G.
[35] Xing Chen, Ming-Xi Liu, and Gui-Ying Yan. Drug–target interaction prediction byrandom walk on the heterogeneous network.Molecular BioSystems, 8(7):1970, jun2012. ISSN 1742-206X. doi: 10.1039/c2mb00002d. URL http://xlink.rsc.org/?DOI=c2mb00002d.
[36] Yu-Fen Huang, Hsiang-Yuan Yeh, and Von-Wun Soo. Inferring drug-disease asso-ciations from integration of chemical, genomic and phenotype data using networkpropagation.BMC medical genomics, 6 Suppl 3(Suppl 3):S4, 2013. ISSN 1755-8794. doi: 10.1186/1755-8794-6-S3-S4. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3980383.
[37] Taehyun Hwang and Rui Kuang. A Heterogeneous Label Propagation Algorithmfor Disease Gene Discovery. Technical report. URL http://www.siam.org/journals/ojsa.php.
[38] Alberto Valdeolivas, Laurent Tichit, Claire Navarro, Sophie Perrin, Ga ̈elle Odelin,Nicolas Levy, Pierre Cau, Elisabeth Remy, and Anas Anas Baudot. Random Walkwith Restart on Multiplex and Heterogeneous Biological Networks. doi: 10.1101/134734. URL http://dx.doi.org/10.1101/134734.48
[39] Maryam Lotfi Shahreza, Nasser Ghadiri, Seyed Rasoul Mousavi, Jaleh Varshosaz,and James R. Green. Heter-LP: A heterogeneous label propagation algorithm and itsapplication in drug repositioning.Journal of Biomedical Informatics, 68:167–183,apr 2017. ISSN 1532-0464. doi: 10.1016/J.JBI.2017.03.006. URL https://www.sciencedirect.com/science/article/pii/S1532046417300552.
[40] A R Aronson. Effective mapping of biomedical text to the UMLS Metathesaurus: theMetaMap program.Proceedings. AMIA Symposium, pages 17–21, 2001. ISSN 1531-605X. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC2243666.
[41] U.S. National Library of Medicine. Pubmed - ncbi. URL https://preview.ncbi.nlm.nih.gov/pubmed.
[42] C E Lipscomb. Medical Subject Headings (MeSH).Bulletin of the Medical Li-brary Association, 88(3):265–6, jul 2000. ISSN 0025-7338. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC35238.
[43] 衛生福利部中醫藥司-基準方劑. URL https://dep.mohw.gov.tw/DOCMAP/lp-768-108.html.
[44] 醫砭-常用方劑. URL http://yibian.hopto.org/fang/.
[45] Yin-Ying Wang, Hong Bai, Run-Zhi Zhang, Hong Yan, Kang Ning, and Xing-Ming Zhao.Predicting new indications of compounds with a network phar-macology approach:Liuwei Dihuang Wan as a case study.Oncotarget, 8(55):93957–93968, nov 2017. doi: 10.1001/archneurpsyc.1948.02300400117008. URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC5706847.
[46] Xujun Liang,Pengfei Zhang,Lu Yan,Ying Fu,Fang Peng,LingzhiQu, Meiying Shao, Yongheng Chen, and Zhuchu Chen.LRSSL: pre-dict and interpret drug–disease associations based on data integration us-ing sparse subspace learning.Bioinformatics,33(8):btw770,jan 2017.ISSN 1367-4803.doi:10.1093/bioinformatics/btw770.URL https://academic.oup.com/bioinformatics/article-lookup/doi/10.1093/bioinformatics/btw770.
[47] Guang-Jian Du, Zhiyu Zhang, Xiao-Dong Wen, Chunhao Yu, Tyler Calway,Chun-Su Yuan, and Chong-Zhi Wang.Epigallocatechin Gallate (EGCG) Isthe Most Effective Cancer Chemopreventive Polyphenol in Green Tea.Nu-trients, 4(11):1679, nov 2012.ISSN 2072-6643.doi: 10.3390/NU4111679.URL http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=PMC3509513.