Julia Meets Mendel: Algorithms and Software for Modern Genomic Data

This repo contains the materials for short course Julia Meets Mendel: Algorithms and Software for Modern Genomic Data at 2018 Joint Statistical Meeting (JSM), Vancouver, Canada.

Instructors

• Janet Sinsheimer, Professor, Biomathematics, Biostatistics, Human Genetics, and Statistics, UCLA

• Eric Sobel, Professor, Human Genetics, UCLA

• Hua Zhou, Associate Professor, Biostatistics, UCLA

Course assistants

• Chris German, PhD student, UCLA

• Sarah Ji, PhD student, UCLA

• Juhyun Kim, PhD student, UCLA

Syllabus

Time	Topic	Instructor
8:30-9:00	participants intro., OpenMendel [pdf]	Eric Sobel
9:00-9:30	Julia: intro. [ipynb], multiple dispatch/JIT/profiling [ipynb]	Hua Zhou
9:30-10:15	Julia: linear algebra [ipynb], optimization [ipynb]	Hua Zhou
10:15-10:30	coffee break
10:30-11:15	Julia: GPU [ipynb], NNMF [ipynb], parallel computing [ipynb]	Hua Zhou
11:15-noon	review of genetic and genomic concepts [pdf]	Janet Sinsheimer
noon-1:00	lunch
1:00-1:30	handling genetic data in OpenMendel [ipynb]	Hua Zhou
1:30-2:30	GWAS for population and genetic analysis [pdf, ipynb]	Eric Sobel
2:30-3:15	linear mixed models for genetic analysis [pdf, ipynb]	Janet Sinsheimer
3:15-3:30	coffee break
3:30-4:30	mendelian randomization [pdf, ipynb]	Janet Sinsheimer
4:30-5:00	Course wrap-up [pdf]	Eric Sobel

Getting started

There are several ways to run Jupyter notebooks in this course.

Run Jupyter notebooks on the dedicated server

The easiest way is to run Jupyter notebooks on a server dedicated to this course.

1. Point your browser to http://35.233.157.190:8000/ (expired 07/31/2018).

2. You'll be greeted by a JupyterHub login page:
   Your username is initial of your first name + your last name. For example, if your name is Joe Bruin, then your username is jbruin. Your password is jsmce18c. Of course you can change password after you log in, e.g., by passwd command at Bash terminal.

3. After login, you will see the JupyterLab interface:

4. To transfer the course materials to your home directory on the server, click Terminal in JupyterLab and type command

git clone https://github.com/Hua-Zhou/JSM2018-CE18C.git

You should see the folder JSM2018-CE18C in the Files tab on the left of JupyterLab. Now you can open any Jupyter notebooks in that folder and run the examples.

5. If you prefer the Jupyter Notebook interface instead of JupyterLab, change the address from http://35.233.157.190:8000/user/jbruin/lab to http://35.233.157.190:8000/user/jbruin/tree.
   

Run Jupyter notebooks on your own laptop

This is not recommended during this short course, since your software environment (OS, Julia version, package versions, etc.) may be quite different from that assumed by the Jupyter notebooks. In case you want to run Jupyter notebooks on your own machine, simply git clone https://github.com/Hua-Zhou/JSM2018-CE18C.git to sync the most recent course materials to your computer and install all needed Julia packages.