Computational Systems Biology: Deep Learning in the Life Sciences
Presents innovative approaches to computational problems in the life sciences, focusing on deep learning-based approaches with comparisons to conventional methods. Topics include protein-DNA interaction, chromatin accessibility, regulatory variant interpretation, medical image understanding, medical record understanding, therapeutic design, and experiment design (the choice and interpretation of interventions). Focuses on machine learning model selection, robustness, and interpretation. Teams complete a multidisciplinary final research project using TensorFlow or other framework. Provides a comprehensive introduction to each life sciences problem, but relies upon students understanding probabilistic problem formulations. Students taking graduate version complete additional assignments.
Spring 2014 · Foundations of Computational and Systems Biology
Topics
- Course Introduction: History of Computational Biology; Overview of the Course; Course Policies and Mechanics; DNA Sequencing Technologies
- Local Alignment (BLAST) and Statistics
- Statistics; Significance Testing; Bonferroni Correction
- Global Alignment of Protein Sequences (NW, SW, PAM, BLOSUM)
- Comparative Genomic Analysis of Gene Regulation
- Clustering, Model Selection, and BIC Scores
- Library Complexity and Short Read Alignment (Mapping)
- Burrows–Wheeler Transform (BWT) and Alignments. Guest Lecture: Heng Li ( Broad Institute )
- Genome Assembly
- ChIP-seq Analysis; DNA-protein Interactions
- Simultaneous ChIP-seq Peak Discovery and Motif Sampling
- RNA-sequence Analysis: Expression, Isoforms
- Modeling and Discovery of Sequence Motifs (Gibbs Sampler, Alternatives)
- Gene Expression Program Discovery Using Topic Models
- Markov and Hidden Markov Models of Genomic and Protein Features
- RNA Secondary Structure—Biological Functions and Prediction
- Probabilistic Grammatical Models of RNA Structure
- Introduction to Protein Structure; Structure Comparison and Classification
- Protein Amino Acid Sidechain Packing Using Markov Random Fields
- Predicting Protein Structure
- Predicting Protein Interactions
- Protein / Protein Interaction Prediction Using Threading
- Gene Regulatory Networks
- Protein Interaction Networks
- Regression Trees
- Logic Modeling of Cell Signaling Networks. Guest Lecture: Doug Lauffenburger
- Analysis of Chromatin Structure
- BayesNets
- Discovering Quantitative Trait Loci (QTLs)
- Narrow Sense Heritability
- Human Genetics, SNPs, and Genome Wide Associate Studies
- Synthetic Biology: From Parts to Modules to Therapeutic Systems. Guest Lecture: Ron Weiss
- Causality, Natural Computing, and Engineering Genomes. Guest Lecture: George Church
- Presentations
- Presentations (cont.)
Readings
- LEC #: 1 · TOPICS: Course Introduction; Overview · READINGS: No readings for this lecture.
- LEC #: 3 · TOPICS: Global Alignment of Protein Statistics · READINGS: No readings for this lecture.
- LEC #: 16 · TOPICS: Protein Interaction Networks · READINGS: No readings for this lecture.
- Leach, Andrew. Molecular Modelling: Principles and Applications . Prentice Hall, 2001. ISBN: 9780582382107.
- Durbin, Richard, Sean R. Eddy, et al. Biological Sequence Analysis: Probabilistic Models of Proteins and Nucleic Acids . Cambridge University Press, 1998. ISBN: 9780521629713.
- Gonick, Larry, and Woollcott Smith. The Cartoon Guide to Statistics . Paw Prints, 2008. ISBN: 9781435242715.
Prerequisites & course information
Subject meets with 6.8710J, 6.8711J, 18.419[J] , 20.390[J] , HST.506[J] Prereq: Biology (GIR) and (6.041 or 18.600 ) G (Spring) Not offered regularly; consult department 3-0-9 units
Sources
- Course catalogue ↗2026–2027
- MIT OpenCourseWare ↗Spring 2014
- MIT OCW readings ↗
- MIT OCW calendar ↗
- MIT OCW syllabus ↗
- Historical MIT OpenCourseWare offering. Named reading entries and schedule topics are automatically extracted from the linked source pages.