Differential_Gene_expression

Table of Contents Objective Procedures Results Conclusion Outlook for the next taks Note and Reference

Objective

  Time estimated: 120 mins; taken 60 mins; date started: 2020-02-27; date completed: 2020-03-11

Procedures

Assignment 2 Part1: Differential Gene Expression

Method: Limma - liner model of microarray analysis (can also apply to RNAseq)

step 1. Define the groups (recall from MDS plot from assignment 1)

step 2. create a data matrix from normalized count data from assignment 1

step 3. Fit the DataMatrix into the linear model

Note: using only cell type VS cell types + patient types

The later is considered as better as it represents more variability in the dataset

step 4. apply empirical Bayes to evaluate differential expression level

Note: set trend=TRUE in eBayes function for RNA-seq data

step 5. compute p values, t values, logFC, average expression level, adj p-value for each gene

step 6. p-value threshold

step 7. correction test

step 8. construct heapmap matrix

Note: construct heatmap for the whole dataset results in PC crashing. To avoid this, a few data cleaning and selection would be needed. example: plot genes with small p-values

Results

Conclusion

Outlook for the next taks

Note and Reference