癌症的高死亡率使其成為不可忽視的重要議題。在癌症治療中，作為精準醫療一環的標靶治療會因癌症的遺傳異質性導致不同患者對治療藥物的反應各不相同，這些差異表現在癌症基因是否產生突變和不同基因表現量上。因此，能夠準確地預測藥物反應至關重要且具有挑戰性。目前基於深度學習的藥物預測方法通過整合癌症相關的多體學資料顯示出良好的性能。然而，這些整合方法存在挑戰，例如在特徵提取後無法將多體學資料有效地投影到相同的潛在空間進行整合，從而導致特徵不對齊。此外，在整合過程中獲取多體學資料之間的相關信息也是一個挑戰。
本研究的目標之一是分析基因相關的多體學資料與癌症之間的關係。我們使用不同癌細胞基因的突變、拷貝數變異、甲基化、基因表現量等多體學資料，預測不同細胞系樣本對藥物的反應濃度。為了解決整合過程中的特徵不對齊問題，本研究提出了特徵空間對齊方法，該方法在特徵提取後將多體學資料對齊到同一特徵空間中。此外，還加入了注意力模組來獲得多體學資料之間的相互作用。最後，探討多體學資料與預測結果的關係，並預期此研究可以為輔助治療藥物選擇的醫療決策奠定基礎。
除了使用多體學的真實數據以外，使用文獻探勘技術也可以協助研究人員於巨量的生物醫學研究資料中擷取特定成果進行分析，例如分析基因與藥物關係。本研究建立了一個生醫文獻探勘模型，藉由基因-特徵矩陣分析基因集特性。在文獻探勘中，EGFR L858R 和T790M，以及BRAF V600E 基因變異是重要的突變特徵，我們也通過不同癌症類型的突變譜對癌症基因面板進行了驗證。在不同的機器學習模型中，對MSK-IMPACT 的預測，最佳準確率分別為 0.959，AUC 分析確認神經網路模型具有最好的預測性能。
另外，在生醫文獻探勘領域中，良好的可解釋性尤為重要，雖然有些機器學習方法有良好的準確率或召回率，但結果通常難以詮釋。因此，我們使用能具有良好可解釋性的主題模型，分析與人類癌症基因有關的文獻。這些實驗結果表明，本研究所建立的生物醫學文獻探勘系統能夠對基因集進行可解釋的分析，並能在臨床研究分析基因集時提供良好的幫助。

Cancer's high mortality rate renders it an imperative issue that demands attention. Targeted therapy in cancer treatment, as part of precision medicine, faces the challenge of genetic heterogeneity in cancer, leading to diverse responses to therapeutic drugs among different patients. These differences are reflected in whether cancer genes undergo mutations and exhibit varying levels of gene expression. Therefore, accurately predicting drug response is crucial and poses a challenge. Current deep learning-based drug prediction methods show good performance by integrating cancer-related multi-omics data. However, these integration methods face challenges, such as the inability to effectively project multi-omics data onto the same latent space after feature extraction, leading to feature mismatch. Additionally, obtaining relevant information and capturing correlations between different omics data during the integration process poses another challenge.
To address these challenges, this study analyzed the relationship between cancer and multi-omics data. Diverse multi-omics data, including mutations, copy number variations, methylation, and gene expression of different cancer cell genes, were used to predict drug response concentrations in distinct samples. This work suggested using latent alignment as a solution to address the problem of information mismatch during integration. Latent alignment involved aligning multi-omics data into a shared latent space after extracting relevant features. Furthermore, an attention module was implemented to effectively capture the interactions among the various forms of omics data. Finally, the relationship between multi-omics data and prediction results were explored, with the expectation that this study will lay the foundation for assisting medical decision making in drug selection for cancer treatment.
In addition to using real multi-omics data, employing literature mining techniques can assist researchers in extracting specific findings from a vast amount of biomedical research data for analysis, such as analyzing the relationships between genes and drugs. This study developed a biomedical literature mining model to analyze gene set characteristics using a gene term-feature matrix. In literature mining, EGFR L858R and T790M, as well as BRAF V600E, were identified as significant mutation features. We also validated the cancer panel through the mutational landscape of various cancer types. Among several machine learning models, the prediction for MSK-IMPACT achieved a maximum accuracy of 0.959, with the area under the receiver operating characteristic curve analysis confirming that the neural net model exhibited the best predictive performance.
Additionally, interpretability is particularly crucial in the domain of biomedical literature mining. While certain machine learning methods achieve high accuracy or recall, their results are often difficult to interpret. To address this, we employed interpretable topic modeling techniques to analyze literature related to human cancer genes. The experimental results demonstrate that the biomedical literature mining system developed in this study provides interpretable analyses of gene sets and offers valuable assistance in clinical research for analyzing gene panels.

1. 統計處 112 年國人死因統計結果 Available online: https://www.mohw.gov.tw/cp-6650-79055-1.html (accessed on 9 July 2024).
2. Stratton, M.R.; Campbell, P.J.; Futreal, P.A. The Cancer Genome. Nature 2009, 458, 719–724,doi:10.1038/nature07943.
3. Paczkowska, M.; Barenboim, J.; Sintupisut, N.; Fox, N.S.; Zhu, H.; Abd-Rabbo, D.; Mee, M.W.; Boutros, P.C.; Reimand, J. Integrative Pathway Enrichment Analysis of Multivariate Omics Data. Nat Commun 2020, 11, 735, doi:10.1038/s41467-019-13983-9.
4. Guan, N.-N.; Zhao, Y.; Wang, C.-C.; Li, J.-Q.; Chen, X.; Piao, X. Anticancer Drug Response Prediction in Cell Lines Using Weighted Graph Regularized Matrix Factorization. Molecular Therapy - Nucleic Acids 2019, 17, 164–174, doi:10.1016/j.omtn.2019.05.017.
5. Liu, P.; Li, H.; Li, S.; Leung, K.-S. Improving Prediction of Phenotypic Drug Response on Cancer Cell Lines Using Deep Convolutional Network. BMC Bioinformatics 2019, 20, 408, doi:10.1186/s12859-019-2910-6.
6. Cheng, D.T.; Mitchell, T.N.; Zehir, A.; Shah, R.H.; Benayed, R.; Syed, A.; Chandramohan, R.; Liu, Z.Y.; Won, H.H.; Scott, S.N.; et al. Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-Based Next-Generation Sequencing Clinical Assay for Solid Tumor Molecular Oncology. J Mol Diagn 2015, 17, 251–264, doi:10.1016/j.jmoldx.2014.12.006.
7. Malik, V.; Kalakoti, Y.; Sundar, D. Deep Learning Assisted Multi-Omics Integration for Survival and Drug-Response Prediction in Breast Cancer. BMC Genomics 2021, 22, 214, doi:10.1186/s12864-021-07524-2.
8. Khan, D.; Shedole, S. Leveraging Deep Learning Techniques and Integrated Omics Data for Tailored Treatment of Breast Cancer. Journal of Personalized Medicine 2022, 12, 674, doi:10.3390/jpm12050674.
9. Chiu, Y.-C.; Chen, H.-I.H.; Zhang, T.; Zhang, S.; Gorthi, A.; Wang, L.-J.; Huang, Y.; Chen, Y. Predicting Drug Response of Tumors from Integrated Genomic Profiles by Deep Neural Networks. BMC Medical Genomics 2019, 12, 18, doi:10.1186/s12920-018-0460-9.
10. Sharifi-Noghabi, H.; Zolotareva, O.; Collins, C.C.; Ester, M. MOLI: Multi-Omics Late Integration with Deep Neural Networks for Drug Response Prediction. Bioinformatics 2019, 35, i501–i509, doi:10.1093/bioinformatics/btz318.
11. Ille, A.M.; Mathews, M.B. AI Interprets the Central Dogma and Genetic Code. Trends in Biochemical Sciences 2023, 48, 1014–1018, doi:10.1016/j.tibs.2023.09.004.
12. Shapiro, J.A. Revisiting the Central Dogma in the 21st Century. Ann N Y Acad Sci 2009, 1178, 6–28, doi:10.1111/j.1749-6632.2009.04990.x.
13. Meyer, U.A.; Zanger, U.M.; Schwab, M. Omics and Drug Response. Annu Rev Pharmacol Toxicol 2013, 53, 475–502, doi:10.1146/annurev-pharmtox-010510-100502.
14. Theodoridis, T.; Chatzis, T.; Solachidis, V.; Dimitropoulos, K.; Daras, P. Cross-Modal Variational Alignment of Latent Spaces. In Proceedings of the 2020 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW); June 2020; pp. 4127–4136.
15. Wang, C.; Lye, X.; Kaalia, R.; Kumar, P.; Rajapakse, J.C. Deep Learning and Multi-Omics Approach to Predict Drug Responses in Cancer. BMC Bioinformatics 2022, 22, 632, doi:10.1186/s12859-022-04964-9.
16. Du, J.; Jia, P.; Dai, Y.; Tao, C.; Zhao, Z.; Zhi, D. Gene2vec: Distributed Representation of Genes Based on Co-Expression. BMC Genomics 2019, 20, 82, doi:10.1186/s12864-018-5370-x.
17. Leaman, R.; Islamaj Doğan, R.; Lu, Z. DNorm: Disease Name Normalization with Pairwise Learning to Rank. Bioinformatics 2013, 29, 2909–2917, doi:10.1093/bioinformatics/btt474.
18. Wei, C.-H.; Kao, H.-Y.; Lu, Z. PubTator: A Web-Based Text Mining Tool for Assisting Biocuration. Nucleic Acids Research 2013, 41, W518–W522, doi:10.1093/nar/gkt441.
19. Wang, Y.; Wu, S.; Li, D.; Mehrabi, S.; Liu, H. A Part-Of-Speech Term Weighting Scheme for Biomedical Information Retrieval. J Biomed Inform 2016, 63, 379–389, doi:10.1016/j.jbi.2016.08.026.
20. Ikonomakis, E.; Kotsiantis, S.; Tampakas, V. Text Classification Using Machine Learning Techniques. WSEAS transactions on computers 2005, 4, 966–974.
21. Devarajan, K.; Wang, G.; Ebrahimi, N. A Unified Statistical Approach to Non-Negative Matrix Factorization and Probabilistic Latent Semantic Indexing. Mach Learn 2015, 99, 137–163, doi:10.1007/s10994-014-5470-z.
22. Choo, J.; Lee, C.; Reddy, C.K.; Park, H. UTOPIAN: User-Driven Topic Modeling Based on Interactive Nonnegative Matrix Factorization. IEEE Trans Vis Comput Graph 2013, 19, 1992–2001, doi:10.1109/TVCG.2013.212.
23. Pedregosa, F.; Varoquaux, G.; Gramfort, A.; Michel, V.; Thirion, B.; Grisel, O.; Blondel, M.; Prettenhofer, P.; Weiss, R.; Dubourg, V.; et al. Scikit-Learn: Machine Learning in Python. J. Mach. Learn. Res. 2011, 12, 2825–2830.
24. SHABANI AZIM, F.; HOURI, H.; GHALAVAND, Z.; NIKMANESH, B. Next Generation Sequencing in Clinical Oncology: Applications, Challenges and Promises: A Review Article. Iran J Public Health 2018, 47, 1453–1457.
25. McCabe, M.J.; Gauthier, M.-E.A.; Chan, C.-L.; Thompson, T.J.; De Sousa, S.M.C.; Puttick, C.; Grady, J.P.; Gayevskiy, V.; Tao, J.; Ying, K.; et al. Development and Validation of a Targeted Gene Sequencing Panel for Application to Disparate Cancers. Sci Rep 2019, 9, 17052, doi:10.1038/s41598-019-52000-3.
26. Hyman, D.M.; Solit, D.B.; Arcila, M.E.; Cheng, D.T.; Sabbatini, P.; Baselga, J.; Berger, M.F.; Ladanyi, M. Precision Medicine at Memorial Sloan Kettering Cancer Center: Clinical next53 Generation Sequencing Enabling next-Generation Targeted Therapy Trials. Drug Discov Today 2015, 20, 1422–1428, doi:10.1016/j.drudis.2015.08.005.
27. Zehir, A.; Benayed, R.; Shah, R.H.; Syed, A.; Middha, S.; Kim, H.R.; Srinivasan, P.; Gao, J.; Chakravarty, D.; Devlin, S.M.; et al. Mutational Landscape of Metastatic Cancer Revealed from Prospective Clinical Sequencing of 10,000 Patients. Nat Med 2017, 23, 703–713, doi:10.1038/nm.4333.
28. Luthra, R.; Patel, K.P.; Routbort, M.J.; Broaddus, R.R.; Yau, J.; Simien, C.; Chen, W.; Hatfield, D.Z.; Medeiros, L.J.; Singh, R.R. A Targeted High-Throughput Next-Generation Sequencing Panel for Clinical Screening of Mutations, Gene Amplifications, and Fusions in Solid Tumors. J Mol Diagn 2017, 19, 255–264, doi:10.1016/j.jmoldx.2016.09.011.
29. Ghandi, M.; Huang, F.W.; Jané-Valbuena, J.; Kryukov, G.V.; Lo, C.C.; McDonald, E.R.; Barretina, J.; Gelfand, E.T.; Bielski, C.M.; Li, H.; et al. Next-Generation Characterization of the Cancer Cell Line Encyclopedia. Nature 2019, 569, 503–508, doi:10.1038/s41586-019-1186-3.
30. Emdadi, A.; Eslahchi, C. DSPLMF: A Method for Cancer Drug Sensitivity Prediction Using a Novel Regularization Approach in Logistic Matrix Factorization. Front Genet 2020, 11, 75, doi:10.3389/fgene.2020.00075.
31. Gillespie, M.; Jassal, B.; Stephan, R.; Milacic, M.; Rothfels, K.; Senff-Ribeiro, A.; Griss, J.; Sevilla, C.; Matthews, L.; Gong, C.; et al. The Reactome Pathway Knowledgebase 2022. Nucleic Acids Res 2022, 50, D687–D692, doi:10.1093/nar/gkab1028.
32. Wei, C.-H.; Kao, H.-Y.; Lu, Z. GNormPlus: An Integrative Approach for Tagging Genes, Gene Families, and Protein Domains. BioMed Research International 2015, 2015, 918710, doi:10.1155/2015/918710.
33. Westlake, A.J. Introduction to Probability Theory and Statistical Inference. Journal of the Royal Statistical Society Series D: The Statistician 1970, 19, 352–353, doi:10.2307/2986843.
34. Yeganova, L.; Kim, W.; Kim, S.; Wilbur, W.J. Retro: Concept-Based Clustering of Biomedical Topical Sets. Bioinformatics 2014, 30, 3240–3248, doi:10.1093/bioinformatics/btu514.
35. Paez, J.G.; Jänne, P.A.; Lee, J.C.; Tracy, S.; Greulich, H.; Gabriel, S.; Herman, P.; Kaye, F.J.; Lindeman, N.; Boggon, T.J.; et al. EGFR Mutations in Lung Cancer: Correlation with Clinical Response to Gefitinib Therapy. Science 2004, 304, 1497–1500, doi:10.1126/science.1099314.
36. Shepherd, F.A.; Rodrigues Pereira, J.; Ciuleanu, T.; Tan, E.H.; Hirsh, V.; Thongprasert, S.; Campos, D.; Maoleekoonpiroj, S.; Smylie, M.; Martins, R.; et al. Erlotinib in Previously Treated Non-Small-Cell Lung Cancer. N Engl J Med 2005, 353, 123–132, doi:10.1056/NEJMoa050753.
37. Zhou, W.; Ercan, D.; Chen, L.; Yun, C.-H.; Li, D.; Capelletti, M.; Cortot, A.B.; Chirieac, L.; Iacob, R.E.; Padera, R.; et al. Novel Mutant-Selective EGFR Kinase Inhibitors against EGFR T790M. Nature 2009, 462, 1070–1074, doi:10.1038/nature08622.
38. Burris, H.A. Dual Kinase Inhibition in the Treatment of Breast Cancer: Initial Experience with the EGFR/ErbB-2 Inhibitor Lapatinib. Oncologist 2004, 9 Suppl 3, 10–15, doi:10.1634/theoncologist.9-suppl_3-10.
39. Arriagada, R.; Bergman, B.; Dunant, A.; Le Chevalier, T.; Pignon, J.-P.; Vansteenkiste, J.; International Adjuvant Lung Cancer Trial Collaborative Group Cisplatin-Based Adjuvant Chemotherapy in Patients with Completely Resected Non-Small-Cell Lung Cancer. N Engl J Med 2004, 350, 351–360, doi:10.1056/NEJMoa031644.
40. Chapman, P.B.; Hauschild, A.; Robert, C.; Haanen, J.B.; Ascierto, P.; Larkin, J.; Dummer, R.; Garbe, C.; Testori, A.; Maio, M.; et al. Improved Survival with Vemurafenib in Melanoma with BRAF V600E Mutation. N Engl J Med 2011, 364, 2507–2516, doi:10.1056/NEJMoa1103782.
41. Hauschild, A.; Grob, J.-J.; Demidov, L.V.; Jouary, T.; Gutzmer, R.; Millward, M.; Rutkowski, P.; Blank, C.U.; Miller, W.H.; Kaempgen, E.; et al. Dabrafenib in BRAF-Mutated Metastatic Melanoma: A Multicentre, Open-Label, Phase 3 Randomised Controlled Trial. Lancet 2012, 380, 358–365, doi:10.1016/S0140-6736(12)60868-X.
42. Seidel, J.A.; Otsuka, A.; Kabashima, K. Anti-PD-1 and Anti-CTLA-4 Therapies in Cancer: Mechanisms of Action, Efficacy, and Limitations. Front Oncol 2018, 8, 86, doi:10.3389/fonc.2018.00086.
43. Rosenbaum, M.W.; Bledsoe, J.R.; Morales-Oyarvide, V.; Huynh, T.G.; Mino-Kenudson, M. PDL1 Expression in Colorectal Cancer Is Associated with Microsatellite Instability, BRAF Mutation, Medullary Morphology and Cytotoxic Tumor-Infiltrating Lymphocytes. Mod Pathol 2016, 29, 1104–1112, doi:10.1038/modpathol.2016.95.
44. Rhodes, D.R.; Kalyana-Sundaram, S.; Mahavisno, V.; Varambally, R.; Yu, J.; Briggs, B.B.; Barrette, T.R.; Anstet, M.J.; Kincead-Beal, C.; Kulkarni, P.; et al. Oncomine 3.0: Genes, Pathways, and Networks in a Collection of 18,000 Cancer Gene Expression Profiles. Neoplasia 2007, 9, 166–180, doi:10.1593/neo.07112.
45. Paige, S.L.; Saha, P.; Priest, J.R. Beyond Gene Panels: Whole Exome Sequencing for Diagnosis of Congenital Heart Disease. Circ Genom Precis Med 2018, 11, e002097, doi:10.1161/CIRCGEN.118.002097.
46. Hochhaus, A.; O’Brien, S.G.; Guilhot, F.; Druker, B.J.; Branford, S.; Foroni, L.; Goldman, J.M.; Müller, M.C.; Radich, J.P.; Rudoltz, M.; et al. Six-Year Follow-up of Patients Receiving Imatinib for the First-Line Treatment of Chronic Myeloid Leukemia. Leukemia 2009, 23, 1054–1061, doi:10.1038/leu.2009.38.
47. Chaouchi, N.; Wallon, C.; Goujard, C.; Tertian, G.; Rudent, A.; Caput, D.; Ferrera, P.; Minty, A.; Vazquez, A.; Delfraissy, J.F. Interleukin-13 Inhibits Interleukin-2-Induced Proliferation and Protects Chronic Lymphocytic Leukemia B Cells from in Vitro Apoptosis. Blood 1996, 87, 1022–1029.
48. Peña-Martínez, P.; Eriksson, M.; Ramakrishnan, R.; Chapellier, M.; Högberg, C.; Orsmark-Pietras, C.; Richter, J.; Andersson, A.; Fioretos, T.; Järås, M. Interleukin 4 Induces Apoptosis of Acute Myeloid Leukemia Cells in a Stat6-Dependent Manner. Leukemia 2018, 32, 588–596, doi:10.1038/leu.2017.261.
49. Martín-Aguilar, A.E.; Núñez-López, H.; Ramirez-Sandoval, J.C. Sorafenib as a Second-Line Treatment in Metastatic Renal Cell Carcinoma in Mexico: A Prospective Cohort Study. BMC Cancer 2021, 21, 16, doi:10.1186/s12885-020-07720-5.
50. Ren, Z.; Niu, Y.; Fan, B.; Wei, S.; Ma, Y.; Zhang, X.; Guo, X.; Zhang, A. Clinical Analysis of Everolimus in the Treatment of Metastatic Renal Cell Carcinoma. Ann Palliat Med 2021, 10, 584–589, doi:10.21037/apm-20-2465.
51. Myszczyszyn, A.; Czarnecka, A.M.; Matak, D.; Szymanski, L.; Lian, F.; Kornakiewicz, A.; Bartnik, E.; Kukwa, W.; Kieda, C.; Szczylik, C. The Role of Hypoxia and Cancer Stem Cells in Renal Cell Carcinoma Pathogenesis. Stem Cell Rev Rep 2015, 11, 919–943, doi:10.1007/s12015-015-9611-y.
52. Truninger, K.; Menigatti, M.; Luz, J.; Russell, A.; Haider, R.; Gebbers, J.-O.; Bannwart, F.; Yurtsever, H.; Neuweiler, J.; Riehle, H.-M.; et al. Immunohistochemical Analysis Reveals High Frequency of PMS2 Defects in Colorectal Cancer. Gastroenterology 2005, 128, 1160–1171, doi:10.1053/j.gastro.2005.01.056.