Skip to content

Latest commit

 

History

History
53 lines (32 loc) · 2.68 KB

FAQs.md

File metadata and controls

53 lines (32 loc) · 2.68 KB

Snippets for Frequently Asked Questions from clients, when they go over the reports.

Feel free to add the FAQs you have received.

General

Functional Analysis

  1. What is geneRatio and bgRatio in overerpresentation analysis?
  • The geneRatio is the {# of annotated genes assigned to term from input}/{# of input genes annotated}

  • The bgRatio is the {# of annotated genes assigned to term from background}/{# of background genes annotated}

Please note that the denominator may be different between MF, BP, and CC as there are different number of genes annotated for those categories.

  • The input is a list of candidate genes (i.e. list of significant DEGs) while the background is a list of all the genes in the study.
  1. How is the p-value calculated?

Simplistic link: https://www.pathwaycommons.org/guide/primers/statistics/fishers_exact_test/

A bit more mathematical link: http://www.nonlinear.com/progenesis/qi/v2.0/faq/should-i-use-enrichment-or-over-representation-analysis-for-pathways-data.aspx

Good video link: https://www.coursera.org/lecture/bd2k-lincs/enrichment-analysis-part-1-xLgN5

Multiple testing correction

What is q-value and why do we need this?

Here is a pretty good slide about the need for multiple testing and how FDR is calculated : https://www.gs.washington.edu/academics/courses/akey/56008/lecture/lecture10.pdf

t-test and Wilcoxon {credits to Preeti}

Blog by Jonathan Bartlett, super helpful for many stat related questions.

https://thestatsgeek.com/2014/04/12/is-the-wilcoxon-mann-whitney-test-a-good-non-parametric-alternative-to-the-t-test/

UpSetR plots

To visualize the overlaps, we use the UpSetR package in R to draw bar plots that demonstrate the overlap, instead of Venn diagrams. The bar plots drawn by this package, and their associated annotations, are a cleaner way to demonstrate/observe overlaps. Here is brief guide to reading the UpSetR overlap plots:

These plots are relatively intuitive for 2 or 3 categories, but can tend to get more complex for >3 categories. In all cases, you will find the categories being compared and their size listed below the bar plots on the left. As you look to the right (directly below each bar) there are dots with connecting lines that denote which categories the overlap is between, or if there is no overlap (just a dot). The numbers at the top of the bars denote the size of the overlap.

PCA

For understanding PCA: Our lesson - https://hbctraining.github.io/scRNA-seq_online/lessons/05_normalization_and_PCA.html#principal-component-analysis-pca A youtube video - https://www.youtube.com/watch?v=_UVHneBUBW0

ChIP-seq and ATAC-seq

BULK RNA-seq

scRNA-seq