BINF G4000 Acculturation to Programming and Statistics

Course Description: This course is targeted for biomedical scientists looking for working knowledge of programming and statistics. This is a fast-paced, hands-on course covering the following topics: programming basics in Python, probabilities, elements of linear algebra, elements of calculus, and elements of data analytics. Students are expected to learn lecture material outside of the classroom and focus on labs during class. All labs evolve around real-world biomedical and health datasets. Only open to DBMI enrolled students in our MA or PhD program. BINF G4000 must be taken fall term of entry. Instructor provides placement exam on first day of class. Students may test out of the course based on placement exam results.

Instructor

Karthik Natarajan, PhD

Class Schedule

This class meets twice weekly on Tuesdays and Thursdays.

Class Structure

I. Programming Basics I (2.5 weeks)

• Computing environment for biomedical sciences (Linux operating system, shell commands)
• IDEs (Integrated Development Environments) and text editors for Python (Emacs, Eclipse)
• Python variables, functions, basic data structures, libraries, numpy, files

Class	Topic	Objectives and Competencies	Themes
00	Introduction to G4000	Overview of course
01	Assessment Test	Assessment Test
02	Introduction to Linux	VirtualBox setup, deploying lubuntu, basic commands	1. ls 2. chmod 3. sudo
03	Introduction to Regex	grep, sed	1. backreference 2. commands 3. UMLS
04	Development environments, programming primitives	Emacs, Python, variables, conditionals, loops, lists	1. keybindings 2. filtering and lists 3. early termination
05	Abstraction of code and data, code reuse	libraries, functions, data structures, files	1. OS standard library 2. lists and tuples 3. recursion
06	Vectorized code and visualization	Vectorized operations and efficiency considerations, reading and writing files, line plots, histograms

II. Probabilities (2 weeks)

• Axioms, PMFs, PDFs, CDFs
• Distribution families (e.g., Binomial, Multinomial, Normal)
• Sampling
• Estimation
• Plotting
• http://research.cs.tamu.edu/prism/lectures/sp/l10.pdf
• http://www.math.uiuc.edu/~kkirkpat/SampleSpace.pdf

Class	Topic	Objectives and competencies	Themes
07	Introduction to probability theory	Axioms, conditional probability law of total probability, sample spaces	1. Sample spaces and events 2. Probability axioms 3. Chain rule of probability
08	Probability distributions	Probability density functions, Probability mass functions, Cumulative distribution functions, mean, variance
09	Random sampling and estimation	Sampling, expected value, MLE, bootstrapping
10	Bayesian probability	Cox’s Theorem, Bayes theorem, interpretations of probability, MAP	1. Cox’s theorem 2. Bayes’ theorem 3. MAP

III. Programming Basics II (1 week)

• Data structures (dictionaries/hash-maps, sets)
• Persistence (reading and writing delimited and JSON files)

Class	Topic	Objectives and Competencies
11	Data structures: dictionaries/hash-maps and sets	Data structure performance characteristics and choice
12	Midterm Review

IV. Elements of Linear Algebra (1 week)

• Scalars, vectors, matrices
• Dot product, matrix multiplication
• Plotting

Class	Topic	Objectives and Competencies
13	Concepts from Linear Algebra	Vectors, matrices, inner product
14	Multidimensional randomness	Random vectors, covariance, multivariate normal distribution

V. Programming Basics III (2 week)

• Persistence (relational database rationale and basic operations)

Class	Topic	Objectives and Competencies	Themes
15	Relational databases and basic operations: Create, Read, Update, and Delete (CRUD)	schema, primary keys, group by	1. LIKE 2. CONCATENATE 3. Functions
16	Database modeling with multiple tables	join, indexes	1. subqueries 2. outer join 3. multi-column index
17	OHDSI
18	Git	version control, git

VI. Programming Basics IV (1 week)

• Object oriented programming
• Handling large datasets (data frames, pandas)

Class	Topic	Objectives and Competencies	Themes
19	Object¬-oriented programming (OOP)	Classes, objects, and inheritance
20	Data, Persistence	Data frames, null values, filtering, strengths and weaknesses of file formats	1. JSON 2. XML 3. CSV 4. Serialization

VII. Elements of Data Analytics (1.5 week)

• Hypothesis testing (scipy library, chi-square, t-test, one-way ANOVA, correlation)
• Predicting from large datasets (logistic regression, convex optimization)

Class	Topic	Objectives and Competencies
21	Hypothesis testing theory	null hypothesis, p-value, confidence interval, credible interval
22	Hypothesis testing practice	chi-squared, t-test, ANOVA, Pearson correlation, non¬parametric tests
23	Prediction	regression, least squares, ML

VIII. Review and Final Exam