Microarray Workshop (2 day)
When: Sep. 29th, 2014 - Sep. 30th, 2014 9:30 am - 4:30 pm
To Know
About this Class
PLEASE NOTE: This 2 day workshop is a BYOC (Bring your own LapTop Computer) class. In order to provide more flexibility with room scheduling we are experimenting with a new format that involves students brining their own laptop computers to the class. This has the advantage that you can continue exactly where you left-off following the class. Government issued or personal computers are permitted. We will be able to supply a very limited set of computers, so if you want to take the class but cannot bring your own computer please indicate such in the Comment section on the registration form.
Direction of FAES Classrooms (B1C204, B1C205) can be found here http://www.faes.org/announcements/directions_faes_classrooms_nih_campus
Day 1 - AM (9:30-12) Introductory Lecture
(Maggie Cam, PhD - CCR, NCI)
Introduction
- Historical Perspective
- Microarray Technologies, Sample Processing Methods
- Microarray comparisons to RNA-Seq
Data Analysis
- Experimental Design
- QC methods
- Preprocessing: Normalization and low level analysis algorithms
Statistical Analysis
- Common statistical models used for analysis of microarray data
- Examples of blocking
- Batch effects and removal methods
Visualization and Clustering
- Volcano Plot
- Principal Components Analysis
- Hierarchical Clustering
- K-means Clustering
Validation and Downstream Analysis
- Validation methods
- Gene Ontology Enrichment and Pathway analysis tools
- Major Software applications
- Public Repositories of Microarray Data
Bioinformatics Core Presentation (Manjula Kasoji - CCRIFX)
- Lessons learned and how to work with the core
Day 1 - PM (1-4:30 pm): Hands-on Microarray analysis using Partek Genomics Suite
(Xiaowen Wang, PhD - Partek)
Partek Genomics Suite Analysis Workflow
- Process Cel files (RMA)
- Looking at data distributions, histograms, bar plots, MA plots, etc.
- Statistical Analysis (Anova)
- Create contrasts
- False Discovery Analysis
- Making lists of significant genes
- Venn Diagrams
Work independently on dataset
Day 2 AM (9:30-12): Hands-on Partek Genomics Suite Analysis and Partek Pathway
(Xiaowen Wang, PhD - Partek)
- Unsupervised Clustering
- Custom Filtering
- Pathway ANOVA
- Work independently on another dataset
Day 2 PM (1-4:30): GeneSet Enrichment Analysis (GSEA)
(Alan Berger, PhD - School of Medicine Johns Hopkins University)
GSEA is a computational method that determines which (if any) a priori defined sets of genes are significantly differentially expressed, as an ensemble, between two biological states. It is an open-source program developed by the Broad Institute: http://www.broadinstitute.org/gsea/index.jsp
Lecture
- The general approach of gene set enrichment methods and comparison with DAVID
- How GSEA measures differential expression for each set of genes
- Controlling effects of multiple comparisons in GSEA (false discovery rate)
- The Broad Institute library of groups of gene sets (MSigDB)
- What files and formats are needed for GSEA
- User options and running GSEA
Hands-on
- Loading the GSEA required input files for an example dataset
- Using and choosing values in the GSEA GUI interface
- Rank-based analysis
- Full dataset analysis
- Understanding the GSEA outputs and judging significance in the results
Work independently on another dataset