Translational Computational Biology

This course provides a comprehensive, project-based exploration of the translational pipeline, from clinical data to therapeutic design. Students will learn and apply computational methods to integrate heterogeneous patient data, infer causal disease mechanisms using network medicine and advanced statistical models, and design novel interventions, including small molecules, gene editing therapies, and cellular immunotherapies. A central theme is the ‘disease-solver' mindset, emphasizing the practical application of these techniques within the ethical and regulatory frameworks that govern modern medicine. The course culminates in a semester-long capstone project where student teams tackle a real-world disease challenge from bench to bedside.

This course is designed to cultivate a "disease-solver" mindset, moving beyond theoretical knowledge to the practical application of computational biology in medicine. It charts a course along the full "bench-to-bedside-and-back" translational pipeline. Students will begin by learning to integrate and harmonize complex, real-world patient data from heterogeneous sources like Electronic Health Records (EHRs) and multi-omics platforms.

Upon successful completion of this course, students will be able to:

• Master the principles of translational science, understanding the complete research continuum from basic discovery to clinical application and back.
• Develop proficiency in integrating and analyzing large-scale, heterogeneous biomedical data to construct comprehensive "digital patient" models.
• Acquire advanced skills in causal inference, applying network-based and statistical methods to identify the molecular drivers of disease.
• Gain practical experience in the computational design of diverse therapeutic modalities, including small molecules, gene therapies, and cellular immunotherapies.
• Navigate the regulatory and ethical landscape of modern medicine, understanding the requirements for bringing a computational therapeutic concept toward clinical reality.
• Synthesize and apply knowledge through a comprehensive, team-based capstone project that addresses a real-world disease problem from end to end.

After successful completion of the modules, students will be able to:

• Recall the key stages of the translational research pipeline (T1-T4) and the "bench-to-bedside-and-back" paradigm.
• Summarize the primary challenges (e.g., semantic, technical, ethical) in integrating EHR and multi-omics data for personalized medicine.
• Interpret the results of a molecular docking simulation, including binding affinity scores and ligand poses.
• Implement a data integration strategy to harmonize and combine features from distinct clinical and molecular datasets into a unified format for analysis.
• Compare and contrast different data fusion strategies (e.g., early, intermediate, and late integration) and justify the selection of a method for a specific biomedical problem.
• Deconstruct a complex clinical problem into a specific, testable causal hypothesis that can be investigated using available genomic and phenotypic data.
• Assess the "druggability" of a potential biological target based on structural features, network properties, and causal evidence.
• Justify the selection of a therapeutic modality (e.g., small molecule vs. gene editing) for a specific disease target by weighing the biological rationale, technical feasibility, and potential regulatory hurdles.
• Design a comprehensive, end-to-end translational research plan for a chosen disease, encompassing data acquisition, causal analysis, therapeutic design, and in silico validation.

• Reference material will be provided on LMS for every module

 This course will exclusively use open-source software and publicly available data and web tools, ensuring accessibility for all students. Students will be expected to install and use the following on their personal computers:

• Programming Environment: Python (via Anaconda distribution) with standard scientific libraries (pandas, NumPy, scikit-learn, Matplotlib) or R with corresponding packages (tidyverse, caret).
• Network Analysis: Cytoscape for network visualization and analysis.
• Molecular Docking: AutoDock Vina and PyRx.
• Web-Based Tools: All required datasets will be from public repositories such as NCBI GEO, dbGaP, and the UK Biobank. An account on the Galaxy server is preferred – use your Plaksha academic email.

Info not available