監督式機器學習方法常被應用在生物醫學關係提取。缺點是需要帶註釋的
訓練樣本資料集，通常由人工花費大量時間及成本創建。遠程監督通過將知識庫與語料庫結合，自動化註釋訓練語料庫。這種方法在生物醫學非常實用，因為許多生物醫學資料庫已提供可供研究的知識庫，但可使用的註釋語料庫卻數量有限。

而基因-基因交互作用可幫助解釋人類複雜性疾病缺失的遺傳率(heritability)，因此本研究主要目的為發展基因-基因交互關係的提取方法。本研究使用KEGG pathway知識庫的基因-基因交互作用資訊，從PubMed摘要中生成訓練樣本集，並使用基於圖核的方法提取基因-基因交互關係。評估結果最好可以達到F-score為0.79。
本研究發展遠程監督方法，可在自動化創建基因-基因交互關係提取的語料庫的能有效減少人工註釋數據所需花費的大量時間成本；而基於圖核的關係提取方法成功應用在基因-基因交互關係提取，期望本研究成果能幫助精準醫療之實現。

Supervised machine learning methods are often used in biomedical relationship extraction. The drawback is the need for annotated datasets of training samples, which are usually created at considerable time and cost by manual. Distant supervised can automatically annotate and train corpus by combining knowledge base with corpus. This approach is useful in biomedicine, where many biomedicine databases already provide a knowledge base to study, but the number of annotated corpora that can be used is limited. 
Gene-gene interaction can help explain heritability of complex diseases in humans, so the main purpose of this study is to develop methods to extract gene-gene interaction.
In this study, gene-gene interaction information from the KEGG pathway knowledge base was used to generate training sample sets from the PubMed abstract, and the gene-gene interaction was extracted by the method based on graph kernel. The best assessment result could be achieved with an F-score of 0.79.
In this study, a distant supervised method is developed, which can effectively reduce the time cost of manually annotating data in automating the creation of gene-gene interaction extracted corpus. The relationship extraction method based on graph kernel has been successfully applied to the extraction of gene-gene interaction relationship. It is expected that the results of this study can help the realization of precision medicine.

目錄
謝誌  I
目錄  I
表目錄  III
圖目錄  IV
第一章 緒論  1
第一節 研究背景  1
第二節 研究動機與目的  2
第二章 文獻探討  3
第一節 生醫文獻探勘  3
第二節 生醫關係提取方法  3
第三節 遠程監督生醫關係提取方法  5
第四節 語料庫  6
2.4.1 現有基因相關語料庫  6
2.4.2 PubMed   8
2.4.3 NER 工具–PubTator   8
2.4.4 GGI 知識庫–KEGG   8
第五節 圖形特徵  9
2.5.1 詞彙特徵(Lexical features)   10
2.5.2 句法特徵(Syntactic features)   10
第六節 基於圖核的分類方法  11
2.6.1 用內核分類  11
2.6.2 圖形內核(Graph Kernel)   13
2.6.3 Python 應用   14
第三章 研究方法  15
第一節 研究架構  15
第二節 資料集  16
3.2.1 KEGG 的參考文獻   16
3.2.2 Mesh 作為關鍵字搜索   16
第三節 文本預處理  17
3.3.1 實體辨識  17
3.3.2 選取基因共現的句子  18
3.3.3 遠程監督語料庫  18
第四節 構造基於特徵的圖形  19
3.4.1 句法依賴解析、詞性標記  19
3.4.2 判斷句子類別  20
3.4.3 基於特徵的圖形  22
第五節 SVM-圖形內核  24
3.5.1. Shortest-Path kernel, SPK   25
3.5.2. Weisfeiler-Lehman Kernel, WLK   26
3.5.3. Neighborhood Hash Kernel, NHK   27
第四章 結果  30
第一節 資料  30
4.1.1. 資料集劃分  30
4.1.2. 分類設置  31
第二節 評估結果  32
4.2.1. CV 結果   32
4.2.2. 分類結果  33
第三節 WLK & NHK 方法比較   36
第四節 研究發現總結  37
第五節 新基因-基因交互作用   38
第六節 研究資料  39
第五章 討論與結論  40
第一節 討論  40
第二節 結論  41
第三節 未來研究發展  41
第六章 附錄  42
參考文獻  46

表目錄
表格 1、基因相關的NER 語料庫   7
表格 2、基因相關的關係語料庫  7
表格 3、使用依賴圖改善生醫關係提取的研究  9
表格 4、生物醫學依存句法剖析器  11
表格 5、基於核方法的生物醫學關係提取研究  13
表格 6、可用圖形內核調查  14
表格 7、python 中應用圖核的函式庫   14
表格 8、圖形特徵  23
表格 9、三種圖形內核  24
表格 10、XOR 和ROT 的範例   28
表格 11、三種方法的性能評估   35


圖目錄
圖 1 流程架構圖  15
圖 2 PubTator Central 在線提供帶有基因註釋的摘要   18
圖 3 遠程監督語料庫  19
圖 4 句子圖形特徵  20
圖 5 二型糖尿病的KEGG pathway   21
圖 6 特徵圖形示意圖  23
圖 7 資料集劃分  31
圖 8 三種方法的CV 訓練過程   33
圖 9 三種方法的混淆矩陣  34
圖 10 三種方法的ROC-AUC  36
圖 11 預測新基因-基因交互作用的效果   38
附圖A- 1 Random Walk kernel - P,R,F-score   42
附圖A- 2 Random Walk kernel - ROC   42
附圖B- 1 負比例(1:10 & 1:100)-CV   43
附圖B- 2 負比例(1:10 & 1:100)-CM   44
附圖B- 3 負比例(1:10 & 1:100)-P,R,F-score   44
附圖B- 4 負比例(1:10 & 1:100)-ROC   45

1.           Andronis, C., et
al., Literature mining, ontologies and
information visualization for drug repurposing. Briefings in
bioinformatics, 2011. 12(4): p.
357-368.

2.           Demner-Fushman,
D., W.W. Chapman, and C.J. McDonald, What
can natural language processing do for clinical decision support? Journal
of biomedical informatics, 2009. 42(5):
p. 760-772.

3.           Van Landeghem, S.,
et al. Extracting protein-protein
interactions from text using rich feature vectors and feature selection. in
3rd International symposium on Semantic
Mining in Biomedicine (SMBM 2008). 2008. Turku Centre for Computer Sciences
(TUCS).

4.           Zheng, W. and C.
Blake, Using distant supervised learning
to identify protein subcellular localizations from full-text scientific
articles. Journal of biomedical informatics, 2015. 57: p. 134-144.

5.           Junge, A. and L.J.
Jensen, CoCoScore: context-aware
co-occurrence scoring for text mining applications using distant supervision.
Bioinformatics, 2020. 36(1): p.
264-271.

6.           Quirk, C. and H.
Poon, Distant supervision for relation
extraction beyond the sentence boundary. arXiv preprint arXiv:1609.04873,
2016.

7.           Panyam, N.C., et
al., Exploiting graph kernels for high
performance biomedical relation extraction. Journal of biomedical
semantics, 2018. 9(1): p. 1-11.

8.           Airola, A., et
al., All-paths graph kernel for
protein-protein interaction extraction with evaluation of cross-corpus
learning. BMC bioinformatics, 2008. 9(11):
p. 1-12.

9.           Kanehisa, M. and
S. Goto, KEGG: kyoto encyclopedia of
genes and genomes. Nucleic acids research, 2000. 28(1): p. 27-30.

10.         Kang, N., et al., Using rule-based natural language processing
to improve disease normalization in biomedical text. Journal of the
American Medical Informatics Association, 2013. 20(5): p. 876-881.

11.         Leaman, R., R.
Islamaj Doğan, and Z. Lu, DNorm: disease
name normalization with pairwise learning to rank. Bioinformatics, 2013. 29(22): p. 2909-2917.

12.         Xiao, S., et al., Gene polymorphism association with type 2
diabetes and related gene-gene and gene-environment interactions in a Uyghur
population. Medical science monitor: international medical journal of
experimental and clinical research, 2016. 22:
p. 474.

13.         Boudjellal, N., et
al., Biomedical relation extraction using
distant supervision. Scientific Programming, 2020. 2020.

14.         Zhao, S., et al., Recent advances in biomedical literature
mining. Briefings in Bioinformatics, 2020.

15.         Zhou, D., D. Zhong,
and Y. He, Biomedical relation
extraction: from binary to complex. Computational and mathematical methods
in medicine, 2014. 2014.

16.         Proux, D., F.
Rechenmann, and L. Julliard. A pragmatic
information extraction strategy for gathering data on genetic interactions.
in Ismb. 2000.

17.         Ono, T., et al., Automated extraction of information on
protein–protein interactions from the biological literature.
Bioinformatics, 2001. 17(2): p.
155-161.

18.         Miwa, M., et al. A rich feature vector for protein-protein
interaction extraction from multiple corpora. in Proceedings of the 2009 conference on empirical methods in natural
language processing. 2009.

19.         Lodhi, H., et al., Text classification using string kernels.
Journal of Machine Learning Research, 2002. 2(Feb): p. 419-444.

20.         Hochreiter, S. and
J. Schmidhuber, Long short-term memory.
Neural computation, 1997. 9(8): p.
1735-1780.

21.         Zhang, Y. and B.
Wallace, A sensitivity analysis of (and
practitioners' guide to) convolutional neural networks for sentence
classification. arXiv preprint arXiv:1510.03820, 2015.

22.         Devlin, J., et al.,
Bert: Pre-training of deep bidirectional
transformers for language understanding. arXiv preprint arXiv:1810.04805,
2018.

23.         Xu, R. and Q. Wang,
A semi-supervised approach to extract
pharmacogenomics-specific drug–gene pairs from biomedical literature for
personalized medicine. Journal of biomedical informatics, 2013. 46(4): p. 585-593.

24.         Mintz, M., et al. Distant supervision for relation extraction
without labeled data. in Proceedings
of the Joint Conference of the 47th Annual Meeting of the ACL and the 4th
International Joint Conference on Natural Language Processing of the AFNLP.
2009.

25.         Quan, C., M. Wang,
and F. Ren, An unsupervised text mining
method for relation extraction from biomedical literature. PloS one, 2014. 9(7): p. e102039.

26.         Shahab, E., A short survey of biomedical relation
extraction techniques. arXiv preprint arXiv:1707.05850, 2017.

27.         Kabiljo, R., A.B.
Clegg, and A.J. Shepherd, A realistic
assessment of methods for extracting gene/protein interactions from free text.
BMC bioinformatics, 2009. 10(1): p.
1-12.

28.         Lee, L.C., F. Horn,
and F.E. Cohen, Automatic extraction of
protein point mutations using a graph bigram association. PLoS Comput Biol,
2007. 3(2): p. e16.

29.         Johansson-Åkhe, I.,
C. Mirabello, and B. Wallner, Predicting
protein-peptide interaction sites using distant protein complexes as structural
templates. Scientific reports, 2019. 9(1):
p. 1-13.

30.         Ravikumar, K., et
al., Literature mining of protein-residue
associations with graph rules learned through distant supervision. Journal
of biomedical semantics, 2012. 3(3):
p. 1-23.

31.         Liu, M., et al. Relation extraction from biomedical
literature with minimal supervision and grouping strategy. in 2014 IEEE International Conference on
Bioinformatics and Biomedicine (BIBM). 2014. IEEE.

32.         Bobić, T., et al. Improving distantly supervised extraction of
drug-drug and protein-protein interactions. in Proceedings of the Joint Workshop on Unsupervised and Semi-Supervised
Learning in NLP. 2012.

33.         Su, P., et al., Using distant supervision to augment
manually annotated data for relation extraction. PloS one, 2019. 14(7): p. e0216913.

34.         Tran, T. and R.
Kavuluru, Distant supervision for
treatment relation extraction by leveraging MeSH subheadings. Artificial
intelligence in medicine, 2019. 98:
p. 18-26.

35.         Kim, J.-D., et al.,
GENIA ontology, in Tech. Rep. TR-NLP-UT-2006-2, Tsujii
Laboratory, University of Tokyo. 2006.

36.         Huang, M.-S., et
al., Biomedical named entity recognition
and linking datasets: survey and our recent development. Briefings in
Bioinformatics, 2020. 21(6): p.
2219-2238.

37.         Kim, J.-D., et al.,
GENIA corpus—a semantically annotated
corpus for bio-textmining. Bioinformatics, 2003. 19(suppl_1): p. i180-i182.

38.         Kim, J.-D., et al. Introduction to the bio-entity recognition
task at JNLPBA. in Proceedings of the
international joint workshop on natural language processing in biomedicine and
its applications. 2004. Citeseer.

39.         Tanabe, L., et al.,
GENETAG: a tagged corpus for gene/protein
named entity recognition. BMC bioinformatics, 2005. 6(1): p. 1-7.

40.         Thomas, P.E., et
al., Challenges in the association of
human single nucleotide polymorphism mentions with unique database identifiers.
BMC bioinformatics, 2011. 12(4): p.
1-18.

41.         Pérez-Pérez, M., et
al., Evaluation of chemical and
gene/protein entity recognition systems at BioCreative V. 5: the CEMP and GPRO
patents tracks. 2017.

42.         Huang, M.-S., et
al., Revised JNLPBA corpus: A revised
version of biomedical NER corpus for relation extraction task. arXiv
preprint arXiv:1901.10219, 2019.

43.         Faessler, E., et
al. ProGene-A large-scale, high-quality
protein-gene annotated benchmark corpus. in Proceedings of the 12th Language Resources and Evaluation Conference.
2020.

44.         Ding, J., et al., Mining MEDLINE: abstracts, sentences, or
phrases?, in Biocomputing 2002.
2001, World Scientific. p. 326-337.

45.         Hakenberg, J., et
al. LLL’05 challenge: Genic interaction
extraction-identification of language patterns based on alignment and finite
state automata. in Proceedings of the
4th Learning Language in Logic workshop (LLL05). 2005.

46.         Pyysalo, S., et
al., BioInfer: a corpus for information
extraction in the biomedical domain. BMC bioinformatics, 2007. 8(1): p. 1-24.

47.         Alex, B., et al. The ITI TXM corpora: Tissue expressions and
protein-protein interactions. in Proceedings
of LREC. 2008. Citeseer.

48.         Kim, J.-D., et al. Overview of BioNLP’09 shared task on event
extraction. in Proceedings of the
BioNLP 2009 workshop companion volume for shared task. 2009.

49.         Buyko, E., E.
Beisswanger, and U. Hahn. The GeneReg
Corpus for Gene Expression Regulation Events-An Overview of the Corpus and its
In-Domain and Out-of-Domain Interoperability. in LREC. 2010.

50.         Kim, J.-D., et al. Overview of genia event task in bionlp
shared task 2011. in Proceedings of
BioNLP shared task 2011 workshop. 2011.

51.         Verspoor, K., et
al., Annotating the biomedical literature
for the human variome. Database, 2013. 2013.

52.         Dogan, R.I., et al.
BioCreative VI Precision Medicine Track:
creating a training corpus for mining protein-protein interactions affected by
mutations. in BioNLP 2017. 2017.

53.         Wang, Y., et al. An Overview of the Active Gene Annotation
Corpus and the BioNLP OST 2019 AGAC Track Tasks. in Proceedings of The 5th Workshop on BioNLP Open Shared Tasks. 2019.

54.         Legrand, J., et
al., PGxCorpus, a manually annotated
corpus for pharmacogenomics. Scientific data, 2020. 7(1): p. 1-13.

55.         Islamaj Doğan, R.,
et al., The BioC-BioGRID corpus: full
text articles annotated for curation of protein–protein and genetic
interactions. Database, 2017. 2017.

56.         Chang, A.A., K.M.
Heskett, and T.M. Davidson, Searching the
literature using medical subject headings versus text word with PubMed. The
Laryngoscope, 2006. 116(2): p.
336-340.

57.         Wei, C.-H., et al.,
PubTator central: automated concept
annotation for biomedical full text articles. Nucleic acids research, 2019.
47(W1): p. W587-W593.

58.         Wei, C.-H., H.-Y.
Kao, and Z. Lu. PubTator: A PubMed-like
interactive curation system for document triage and literature curation. in
Proceedings of the BioCreative 2012
Workshop, Washington, DC. 2012.

59.         Wei, C.-H., H.-Y.
Kao, and Z. Lu, PubTator: a web-based
text mining tool for assisting biocuration. Nucleic acids research, 2013. 41(W1): p. W518-W522.

60.         Clegg, A.B. and
A.J. Shepherd, Benchmarking
natural-language parsers for biological applications using dependency graphs.
BMC bioinformatics, 2007. 8(1): p.
1-17.

61.         Miyao, Y., et al., Evaluating contributions of natural language
parsers to protein–protein interaction extraction. Bioinformatics, 2009. 25(3): p. 394-400.

62.         Buyko, E. and U.
Hahn. Evaluating the impact of
alternative dependency graph encodings on solving event extraction tasks.
in Proceedings of the 2010 Conference on
Empirical Methods in Natural Language Processing. 2010.

63.         Neumann, M., et
al., Scispacy: Fast and robust models for
biomedical natural language processing. arXiv preprint arXiv:1902.07669,
2019.

64.         Nguyen, D.Q. and K.
Verspoor, From POS tagging to dependency
parsing for biomedical event extraction. BMC bioinformatics, 2019. 20(1): p. 1-13.

65.         Schuster, S., et
al. Paris and Stanford at EPE 2017:
Downstream evaluation of graph-based dependency representations. in EPE 2017-The First Shared Task on Extrinsic
Parser Evaluation. 2017.

66.         Charniak, E. and M.
Johnson. Coarse-to-fine n-best parsing
and maxent discriminative reranking. in Proceedings
of the 43rd Annual Meeting of the Association for Computational Linguistics
(ACL’05). 2005.

67.         Choi, J.D. and A.
McCallum. Transition-based dependency
parsing with selectional branching. in Proceedings
of the 51st Annual Meeting of the Association for Computational Linguistics
(Volume 1: Long Papers). 2013.

68.         Chen, D. and C.D.
Manning. A fast and accurate dependency
parser using neural networks. in Proceedings
of the 2014 conference on empirical methods in natural language processing
(EMNLP). 2014.

69.         Dozat, T., P. Qi,
and C.D. Manning. Stanford’s graph-based
neural dependency parser at the conll 2017 shared task. in Proceedings of the CoNLL 2017 Shared Task:
Multilingual Parsing from Raw Text to Universal Dependencies. 2017.

70.         Nguyen, D.Q., M.
Dras, and M. Johnson, A novel neural
network model for joint POS tagging and graph-based dependency parsing.
arXiv preprint arXiv:1705.05952, 2017.

71.         Kanerva, J., et al.
Turku neural parser pipeline: An
end-to-end system for the CoNLL 2018 shared task. in Proceedings of the CoNLL 2018 Shared Task: Multilingual parsing from
raw text to universal dependencies. 2018.

72.         Beltagy, I., K. Lo,
and A. Cohan, SciBERT: A pretrained
language model for scientific text. arXiv preprint arXiv:1903.10676, 2019.

73.         Dozat, T. and C.D.
Manning, Deep biaffine attention for
neural dependency parsing. arXiv preprint arXiv:1611.01734, 2016.

74.         Giuliano, C., A.
Lavelli, and L. Romano. Exploiting
shallow linguistic information for relation extraction from biomedical
literature. in 11th Conference of the
European Chapter of the Association for Computational Linguistics. 2006.

75.         Kim, S., J. Yoon,
and J. Yang, Kernel approaches for genic
interaction extraction. Bioinformatics, 2008. 24(1): p. 118-126.

76.         Simões, G., H.
Galhardas, and D. Matos, A Labeled Graph
Kernel for Relationship Extraction. arXiv preprint arXiv:1302.4874, 2013.

77.         Cortes, C. and V.
Vapnik, Support-vector networks.
Machine learning, 1995. 20(3): p.
273-297.

78.         Schölkopf, B., A.J.
Smola, and F. Bach, Learning with
kernels: support vector machines, regularization, optimization, and beyond.
2002: MIT press.

79.         Shawe-Taylor, J.
and N. Cristianini, Kernel methods for
pattern analysis. 2004: Cambridge university press.

80.         Zelenko, D., C.
Aone, and A. Richardella, Kernel methods
for relation extraction. Journal of machine learning research, 2003. 3(Feb): p. 1083-1106.

81.         Zhang, Y., et al., Neighborhood hash graph kernel for
protein–protein interaction extraction. Journal of biomedical informatics,
2011. 44(6): p. 1086-1092.

82.         Kriege, N.M., F.D.
Johansson, and C. Morris, A survey on
graph kernels. Applied Network Science, 2020. 5(1): p. 1-42.

83.         Gärtner, T., P.
Flach, and S. Wrobel, On graph kernels:
Hardness results and efficient alternatives, in Learning theory and kernel machines. 2003, Springer. p. 129-143.

84.         Kashima, H., K.
Tsuda, and A. Inokuchi. Marginalized
kernels between labeled graphs. in Proceedings
of the 20th international conference on machine learning (ICML-03). 2003.

85.         Borgwardt, K.M. and
H.-P. Kriegel. Shortest-path kernels on
graphs. in Fifth IEEE international
conference on data mining (ICDM'05). 2005. IEEE.

86.         Shervashidze, N.,
et al. Efficient graphlet kernels for
large graph comparison. in Artificial
intelligence and statistics. 2009. PMLR.

87.         Hido, S. and H.
Kashima. A linear-time graph kernel.
in 2009 Ninth IEEE International Conference
on Data Mining. 2009. IEEE.

88.         Costa, F. and K. De
Grave. Fast neighborhood subgraph
pairwise distance kernel. in ICML.
2010.

89.         Zhang, Y., et al., Hash subgraph pairwise kernel for
protein-protein interaction extraction. IEEE/ACM transactions on computational
biology and bioinformatics, 2012. 9(4):
p. 1190-1202.

90.         Shervashidze, N.,
et al., Weisfeiler-lehman graph kernels.
Journal of Machine Learning Research, 2011. 12(9).

91.         Da San Martino, G.,
N. Navarin, and A. Sperduti. A memory
efficient graph kernel. in The 2012
International Joint Conference on Neural Networks (IJCNN). 2012. IEEE.

92.         Da San Martino, G.,
N. Navarin, and A. Sperduti. A tree-based
kernel for graphs. in Proceedings of
the 2012 SIAM International Conference on Data Mining. 2012. SIAM.

93.         Kriege, N.M., P.-L.
Giscard, and R.C. Wilson, On valid
optimal assignment kernels and applications to graph classification. arXiv
preprint arXiv:1606.01141, 2016.

94.         Neumann, M., et
al., Propagation kernels: efficient graph
kernels from propagated information. Machine Learning, 2016. 102(2): p. 209-245.

95.         Nikolentzos, G., P.
Meladianos, and M. Vazirgiannis. Matching
node embeddings for graph similarity. in Proceedings of the AAAI Conference on Artificial Intelligence.
2017.

96.         Siglidis, G., et
al., GraKeL: A Graph Kernel Library in
Python. Journal of Machine Learning Research, 2020. 21(54): p. 1-5.

97.         Hagberg, A., P.
Swart, and D. S Chult, Exploring network
structure, dynamics, and function using NetworkX. 2008, Los Alamos National
Lab.(LANL), Los Alamos, NM (United States).

98.         Bunescu, R. and R.
Mooney. A shortest path dependency kernel
for relation extraction. in Proceedings
of Human Language Technology Conference and Conference on Empirical Methods in
Natural Language Processing. 2005.

99.         Hendrickx, I., et
al., Semeval-2010 task 8: Multi-way
classification of semantic relations between pairs of nominals. arXiv
preprint arXiv:1911.10422, 2019.

100.      Borgwardt, K., et
al., Graph Kernels: State-of-the-Art and
Future Challenges. arXiv preprint arXiv:2011.03854, 2020.

101.      Dijkstra, E.W., A note on two problems in connexion with
graphs. Numerische mathematik, 1959. 1(1):
p. 269-271.

102.      Weisfeiler, B. and A.
Leman, The reduction of a graph to
canonical form and the algebra which appears therein. NTI, Series, 1968. 2(9): p. 12-16.

103.      Sugiyama, M. and K.
Borgwardt, Halting in random walk
kernels. Advances in neural information processing systems, 2015. 28: p. 1639-1647.

104. Salzberg, S.L., On comparing classifiers: Pitfalls to avoid and a recommended approach. Data mining and knowledge discovery, 1997. 1(3): p. 317-328.