Estimating data structure embedded in semi-structured data

• Project/Area Number: 24300054
• Research Category: Grant-in-Aid for Scientific Research (B)
• Allocation Type: Partial Multi-year Fund
• Section: 一般
• Research Field: Intelligent informatics
• Research Institution: Kyoto University
• Principal Investigator: Mamitsuka Hiroshi 京都大学, 化学研究所, 教授 (00346107)
• Co-Investigator(Renkei-kenkyūsha): Takigawa Ichigaku 北海道大学, 情報科学研究科, 准教授 (10374597) ; Hancock Timothy 京都大学, 化学研究所, 助教 (80600709) ; Shiga Motoki 岐阜大学, 工学部, 助教 (20437263) ; Tsuda Koji 東京大学, 新領域創成科学研究科, 教授 (90357517) ; Kayano Mitsunori 帯広畜産大学, グローバルアグロメディシン研究センター, 講師 (20590095) ; Nguyen Canh Hao 京都大学, 化学研究所, 助教 (90626889)
• Project Period (FY): 2012-04-01 – 2016-03-31
• Project Status: Completed (Fiscal Year 2015)
• Budget Amount: ¥17,940,000 (Direct Cost: ¥13,800,000、Indirect Cost: ¥4,140,000); Fiscal Year 2014: ¥6,110,000 (Direct Cost: ¥4,700,000、Indirect Cost: ¥1,410,000); Fiscal Year 2013: ¥6,500,000 (Direct Cost: ¥5,000,000、Indirect Cost: ¥1,500,000); Fiscal Year 2012: ¥5,330,000 (Direct Cost: ¥4,100,000、Indirect Cost: ¥1,230,000)
• Keywords: 知識発見とデータマイニング / グラフマイニング / 機械学習 / 生体生命情報学 / 人工知能 / ゲノム / プロテオーム

Outline of Final Research Achievements

The objective of this research is to build machine learning solutions for a variety of problems of semi-structured data, particularly graphs and networks. Particular problem focus was "label propagation" and "link prediction". We have not only built machine learning techniques but also applied our technique to real data, particularly those in life sciences.

Research Products

• [Int'l Joint Research] Lille 1 University(France)
• [Int'l Joint Research] Fudan University(中国)
• [Journal Article] Mining approximate patterns with frequent locally optimal occurrences (2016)
  • Author(s): Nakamura A, Takigawa I, Tosaka H, Kudo M, Mamitsuka H
  • Journal Title: Discrete Applied Mathematics Volume: 200 Pages: 123-152
  • DOI: 10.1016/j.dam.2015.07.002
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Current Status and Prospects of Computational Resources for Natural Product Dereplication: A Review. (2016)
  • Author(s): Mohamed, A., Nguyen, C. H. and Mamitsuka, H.
  • Journal Title: Briefings in Bioinformatics Volume: 17 Issue: 2 Pages: 309-321
  • DOI: 10.1093/bib/bbv042
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Non-negative Matrix Factorization with Auxiliary Information on Overlapping Groups (2015)
  • Author(s): Motoki Shiga, Hiroshi Mamitsuka
  • Journal Title: IEEE Transactions on Knowledge and Data Engineering Volume: 27 Issue: 6 Pages: 1615-1628
  • DOI: 10.1109/tkde.2014.2373361
  • NAID: 120005663300
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] MeSHLabeler: Improving the Accuracy of Large-scale MeSH indexing by Integrating Diverse Evidence. (2015)
  • Author(s): Liu, K., Peng, S., Wu, J., Zhai, C., Mamitsuka H. and Zhu S.
  • Journal Title: Bioinformatics Volume: 31 Issue: 12 Pages: i339-i347
  • DOI: 10.1093/bioinformatics/btv237
  • NAID: 120005980853
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] BMExpert: Mining MEDLINE for Finding Experts in Biomedical Domains Based on Language Model. (2015)
  • Author(s): Wang, B., Chen, X., Mamitsuka, H. and Zhu, S.
  • Journal Title: IEEE/ACM Transactions on Computational Biology and Bioinformatics Volume: 12 Issue: 6 Pages: 1286-1294
  • DOI: 10.1109/tcbb.2015.2430338
  • NAID: 120006654074
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] MeSHSim: An R/Bioconductor Package for Measuring Semantic Similarity over MeSH Headings and MEDLINE Documents (2015)
  • Author(s): Zhou, J., Shui, Y., Peng, S., Li, X., Mamitsuka, H. and Zhu, S.
  • Journal Title: Journal of Bioinformatics and Computational Biology Volume: 13 Issue: 06 Pages: 1542002-1542002
  • DOI: 10.1142/s0219720015420020
  • NAID: 120006654075
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Evaluation of Serum-based Cancer Biomarkers: A Brief Review from a Clinical and Computational Viewpoint (2015)
  • Author(s): Yotsukura, S. and Mamitsuka, H.
  • Journal Title: Critical Reviews in Oncology/Hematology Volume: 93 Issue: 2 Pages: 103-115
  • DOI: 10.1016/j.critrevonc.2014.10.002
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] NetPathMiner:R/Bioconductor Package for Network Path Mining through Gene Expression (2014)
  • Author(s): Mohamed, A., Hancock, T., Nguyen, C. H. and Mamitsuka, H.
  • Journal Title: Bioinformatics Volume: 30 Issue: 21 Pages: 3139-3141
  • DOI: 10.1093/bioinformatics/btu501
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Detecting Differentially Coexpressed Genes from Labeled Expression Data:A Brief Review (2014)
  • Author(s): Kayano, M., Shiga, M. and Mamitsuka, H.
  • Journal Title: IEEE/ACM Transactions on Computational Biology and Bioinformatics Volume: 11 Issue: 1 Pages: 154-167
  • DOI: 10.1109/tcbb.2013.2297921
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Selecting Graph Cut Solutions via Global Graph Similarity (2014)
  • Author(s): Nguyen, C. H., Wicker, N. and Mamitsuka, H.
  • Journal Title: IEEE Transactions on Neural Networks and Learning Systems Volume: 25 Issue: 7 Pages: 1407-1412
  • DOI: 10.1109/tnnls.2013.2292975
  • Peer Reviewed / Acknowledgement Compliant
• [Journal Article] Similarity-based machine learning methods for predicting drug.target interactions : a brief review (2013)
  • Author(s): Ding H, Takigawa I, Mamitsuka H, Zhu S
  • Journal Title: Briefings in Bioinformatics Volume: (印刷中・先行オンライン出版済) Issue: 5 Pages: 737-747
  • DOI: 10.1093/bib/bbt056
  • Peer Reviewed
• [Journal Article] Multiple Graph Label Propagation by Sparse Integration (2013)
  • Author(s): Karasuyama, M. and Mamitsuka, H.
  • Journal Title: IEEE Transactions on Neural Networks and Learning Systems Volume: 24 Issue: 12 Pages: 1999-2012
  • DOI: 10.1109/tnnls.2013.2271327
  • Peer Reviewed
• [Journal Article] A New Dissimilarity Measure for Comparing Labeled Graphs (2013)
  • Author(s): Wicker, N., Nguyen, C. H. and Mamitsuka, H.
  • Journal Title: Linear Algebra and its Applications Volume: 438 Issue: 5 Pages: 2331-2338
  • DOI: 10.1016/j.laa.2012.10.021
  • Peer Reviewed
• [Journal Article] Efficient Semi-supervised MEDLINE Document Clustering with MeSH Semantic and Global Content Constraints (2013)
  • Author(s): Gu, J., Feng, W., Zeng, J., Mamitsuka, H. and Zhu, S.
  • Journal Title: IEEE Transactions on Cybernetics Volume: 43 Issue: 4 Pages: 1265-1276
  • DOI: 10.1109/tsmcb.2012.2227998
  • Peer Reviewed
• [Journal Article] Fast Algorithms for Finding a Minimum Repetition Representation of Strings and Trees (2013)
  • Author(s): Nakamura, A., Saito, T., Takigawa, I., Kudo, M. and Mamitsuka, H.
  • Journal Title: Discrete Applied Mathematics Volume: 161 Issue: 10-11 Pages: 1556-1575
  • DOI: 10.1016/j.dam.2012.12.013
  • Peer Reviewed
• [Journal Article] Graph mining: procedure, application to drug discovery and recent advances (2013)
  • Author(s): Ichigaku Takigawa, Hiroshi Mamitsuka
  • Journal Title: Drug Discovery Today Volume: 18: 1-2 Issue: 1-2 Pages: 50-57
  • DOI: 10.1016/j.drudis.2012.07.016
  • Peer Reviewed
• [Journal Article] SIBIC : A Web Server for Generating Gene Set Seiworks Based on Blelusters Obtained by Maximal Frequent Itemsel Mining (2013)
  • Author(s): Keiichio Takahashi. Ichigaku Takigawa and Hiroshi Mamitsuka
  • Journal Title: PLOS ONE Volume: 8(12) Issue: 12 Pages: e82890-e82890
  • DOI: 10.1371/journal.pone.0082890
  • Peer Reviewed
• [Journal Article] An In Silico Model for Interpreting Polypharmacological Relationships in Drug-Target Networks (2013)
  • Author(s): Ichigaku Takigawa, Koji Tsuda, Hiroshi Mamitsuka
  • Journal Title: Methods in Molecular Biology (In Silico Models for Drug Discovery) Volume: 993 Pages: 67-80
  • DOI: 10.1007/978-1-62703-342-8_5
  • ISBN: 9781627033411, 9781627033428
  • Peer Reviewed
• [Journal Article] Identifying Pathways of Coordinated Gene Expression (2013)
  • Author(s): Timothy Hancock, Ichigaku Takigawa, Hiroshi Mamitsuka
  • Journal Title: Methods in Molecular Biology (Data Mining for Systems Biology) Volume: 939 Pages: 69-85
  • DOI: 10.1007/978-1-62703-107-3_7
  • ISBN: 9781627031066, 9781627031073
  • Peer Reviewed
• [Journal Article] Latent Feature Kernels for Link Prediction on Sparse Graphs (2012)
  • Author(s): Nguhen, C. H. and Mamitsuka, H.
  • Journal Title: LEEE Transactions on Neural Networks and Learning Systems Volume: 23 Issue: 11 Pages: 1793-1804
  • DOI: 10.1109/tnnls.2012.2215337
  • Peer Reviewed
• [Journal Article] Boosted Network Classifiers for Local Feature Selection (2012)
  • Author(s): Hancock. T. and Mami tsuka, H.
  • Journal Title: IEEE Transactions on Neural Networks and Learning Systels Volume: 23 Issue: 11 Pages: 1767-1778
  • DOI: 10.1109/tnnls.2012.2214057
  • Peer Reviewed
• [Journal Article] A Variational Bayesian Framework for Clustering with Multiple Graphs (2012)
  • Author(s): Shiga, M. and Mam i tsuka, H.
  • Journal Title: IEEE Transactions on Knowledge and Data Engineering Volume: 24 Issue: 4 Pages: 577-590
  • DOI: 10.1109/tkde.2010.272
  • NAID: 120004007455
  • Peer Reviewed
• [Presentation] Collaborative matrix factorization for predicting drug-target interactions (2015)
  • Author(s): Mamitsuka, H.
  • Organizer: CIM Workshop on Machine Learning
  • Place of Presentation: Uppsala, Sweden
  • Year and Date: 2015-10-08
  • Invited
• [Presentation] Instance-wise Weighted Nonnegative Matrix Factorization for Aggregating Partitions with Locally Reliable Clusters. (2015)
  • Author(s): Zheng, X., Zhu, S., Gao, J. and Mamitsuka, H.
  • Organizer: 24th International Joint Conference on Artificial Intelligence (IJCAI 2015)
  • Place of Presentation: Buenos Aires, Argentina
  • Year and Date: 2015-07-25
• [Presentation] MeSHLabeler: Improving the Accuracy of Large-scale MeSH indexing by Integrating Diverse Evidence (2015)
  • Author(s): Liu, K., Peng, S., Wu, J., Zhai, C., Mamitsuka H. and Zhu S.
  • Organizer: 23rd International Conference on Intelligent Systems for Molecular Biology (ISMB/ECCB 2015)
  • Place of Presentation: Dublin, Ireland
  • Year and Date: 2015-07-10
• [Presentation] Learning kernel-based feature representation for gene essentiality prediction (2014)
  • Author(s): Karasuyama, M. and Mamitsuka, H.
  • Organizer: DREAM Challenges and Cytoscape Workshops
  • Place of Presentation: San Diego, CA, USA
  • Year and Date: 2014-11-09 – 2014-11-14
• [Presentation] Collaborative Matrix Factorization for Predicting Drug-Target Interactions (2014)
  • Author(s): Mamitsuka, H.
  • Organizer: Trends in Machine Learning
  • Place of Presentation: Kyoto University, Japan
  • Invited
• [Presentation] Manifold-based Similarity Adaptation for Label Propagation (2013)
  • Author(s): Karasuyama, M. and Mamitsuka, H.
  • Organizer: Twenty-Seventh Annual Conference on Neural Information Processing Systems (NIPS 2013)
  • Place of Presentation: Lake Tahoe, NV, USA
• [Presentation] Collaborative Matrix Factorization with Multiple Similarities for Predicting Drug-Target Interactions (2013)
  • Author(s): Zheng, X., Ding, H., Mamitsuka, H. and Zhu, S.
  • Organizer: Nineteenth ACM SIGKDD International Conference On Knowledge Discovery and Data Mining (KDD 2013)
  • Place of Presentation: Chicago, IL, USA
• [Presentation] Variational Bayes Co-clustering with Auxiliary Information (2013)
  • Author(s): Shiga, M. and Mamitsuka, H.
  • Organizer: 4th MultiClust Workshop on Multiple Clusterings, Multi-view Data, and Multi-source Knowledge-driven Clustering (MultiClust 2013)
  • Place of Presentation: Chicago, IL, USA
• [Presentation] Machine Learning for Bioinformatics (2013)
  • Author(s): Mamitsuka, H.
  • Organizer: FASEB Science and Research Conference, The Biology of Calpains in Health and Disease
  • Place of Presentation: Saxtons River, VT, USA
  • Invited
• [Presentation] Mining Significant Substructure Pairs from Drug-Target Network (2013)
  • Author(s): 馬見塚 拓
  • Organizer: 日本バイオインフォマティクス学会第１７回創薬インフォマティクス研究会
  • Place of Presentation: 産業技術総合研究所臨海副都心センター別館
  • Invited
• [Presentation] Mining Metabolic Pathways through Gene Expression (2013)
  • Author(s): Mamitsuka, H.
  • Organizer: Kyoto University--Koc University Symposium ``New Frontiers in Health Sciences & Technologies''
  • Place of Presentation: Kyoto University, Japan
  • Invited
• [Presentation] Efficiently Detecting Switching Mechanisms in Gene Expression (2013)
  • Author(s): Mamitsuka, H.
  • Organizer: 23rd International Conference on Genome Informatics (GIW 2013)
  • Place of Presentation: Matrix, Biopolis, Singpore
  • Invited
• [Book] Data Mining for Systems Biology: Methods and Protocols (2013)
  • Author(s): Mamitsuka, H., DeLisi, C., and Kanehisa, M.
  • Total Pages: 279
  • Publisher: Springer
• [Remarks] DREAMチャレンジ 遺伝子エッセンシャリティー予測で第1位を達成
  • URL: http://www.kuicr.kyoto-u.ac.jp/announce/2014/news_141114.html