Translational Bioinformatics in R

Date: Aug 10-14, 2020
Five-day workshop, primarily in R, occasionally in Python
Morning (9-12pm): tutorial (required) (Thursday starts at 8:00am)
Afternoon (1pm - 3pm): lab & QA (optional)

Table of Contents

• Instructors & TAs
• Sessions
  • Introduction to Big Data
  • Data manipulation and visualization
  • Basic statistical analysis
  • Machine learning
  • RNA-Seq
  • Single cell RNA-Seq
  • From big data to drug discovery
  • Structure-based drug discovery
  • R markdown, R packages, R Shiny
  • Q&A, feedback

Instructors & TAs

• Bin Chen
• Ke Liu
• Jing Xing
• Eugene Chekalin
• Mengying Sun
• Jiayu Zhou (invited)
• Yuehua Cui (invited)
• Paul Egeler (invited)
• Eric Kort (invited)

Sessions

Introduction to Big Data

(two sessions, Bin), Day 1: 9-11am

1. Data structure
   • Matrix (data.frame, matrix)
   • Network (igraph, cytoscape, stringdb, KEGG)
   • Unstructured data (text, clinical note, pattern regular expression)
2. Data modality
   • Omics data (genomics, transcriptomics, proteomics, metabolomics)
   • Screening data (CRISPR, pharmacogenomics)
   • Image data (Morphology image)
   • Knowledge graph (PPI)
   • EMR
3. Models
   • cells/organoids/animal models/patients
4. Public data resources
5. Basic R (read/write, data types)
6. Workshop Structure

Data manipulation and visualization

(two sessions, Eugene), Day 1: 11-12pm, Day 2: 9-10am

1. Apply (lapply, sapply, by)
2. Collapse/Summarize data (dplyr)
3. PCA/TSNE/umap (Variation)
4. Clustering (hclust)
5. ggplot2 (boxplot, violin plot, scatter plot, error bar)
6. Heatmap (annotation, clustering, ComplexHeatmap )
7. Publication-ready figures (font, background, legend)

Basic statistical analysis

(two sessions, Yuehua) Day 2: 10-12pm

1. Correlation analysis
2. Continuous data (spearman, pearson)
3. Categorical data (fisher test, chi square)
4. Linear regression and logistic regression (odds ratio)
5. Confounding factors
6. P value and FDR (p value correlation)
7. Survival analysis (KM plot, Hazard Ratio)

Machine learning

(two sessions, Jiayu) Day 3: 9-11am

1. Intro to ML
2. Perceptron and Deep Learning
3. Tree Methods
4. Random Forest
5. Adaboost
6. Gradient Boosting

RNA-Seq

(two sessions, Ke) Day 3: 11-12, Day 4: 8-9am

1. Sequence alignment
2. DE analysis (edgeR, DESeq, Limma Voom)
3. Enrichment analysis (GSEA, ssGSEA, EnrichR, Hypergeometric test, Pathway Databases)

Single cell RNA-Seq

(two sessions, Eric) Day 4: 9-11am

1. Basic (biological theory, common platforms)
2. Sequence analysis (alignment, counting, QC, normalization)
3. Down-stream analysis (dimension reduction, visualization, cell type assignment, and RNA velocity)

From big data to drug discovery

(one session, Bin) Day 4: 11-12

1. Structure representation (SMILES, SD)
2. Structure embedding (fingerprint, pharmacophore)
3. Drug-induced gene expression profiles
4. Systems-based drug discovery (OCTAD)

Structure-based drug discovery

(one session, Jing) Day 5: 9-10am

1. Protein structure/3D structure
2. Docking
3. Target fishing

R markdown, R packages, R Shiny

(one session, Paul) Day 5: 10-11am

Q&A, feedback

(Bin) Day 5: 11:00-12pm