Single-cell RNA-seq analysis with Seurat package

Timothy Tickle and Brian Haas
October 1, 2015

Workshop material available on GitHub

Genes Have Different Distributions

professor corgi

Corner function

# Only the corner
# The full data will be too large to see
library(useful)
corner( data.set )

              ES_A01 ES_B01 ES_C01 ES_D01 ES_E01
Tor1aip2           1      0      1     26      4
Pnkd               0      0     27      1      0
Smyd3              0      0     37      3      0
4921521F21Rik      0      0      0      0      0
Gpbar1             0      0      0      0      0

Seurat: R package

suerat

We Start with a Matrix of Expression

starting_matrix

QC: Viewing Specific Genes in Data

  • Check the identity of the cells with “positive control” genes!!!

plot of chunk unnamed-chunk-3

Viewing Genes vs Genes

(also cells vs cells)

plot of chunk unnamed-chunk-4

t-SNE: Collapsing the Visualization to 2D in non-linear way

tsne_collapsed

t-SNE using Seurat

plot of chunk unnamed-chunk-5

t-SNE: PCA & t-SNE side by side

plot of chunk unnamed-chunk-6

plot of chunk unnamed-chunk-7

SCDE: Single Cell Differential Expression

scde_intro

RaceID: Detecting Rare Cell Populations

raceid_pubs

Remarks

  • data for sale to test this pipeline?
  • Robrecht's tool SCORPIUS: find chronological development in SC data