Skip to content

rajewsky-lab/Eskrate

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

8 Commits
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Eskrate

Eskrate estimates time-dependent RNA kinetic rates using:

  1. single cells with assigned times in biological process
  2. newly synthesized (labeled) and pre-existing (unlabeled) RNAs gene expression data for single cells (e.g. SLAM-Drop-seq data).

To use this package, first install it in R:

devtools::install_github("rajewsky-lab/Eskrate")  

Here's a simple tutorial for estiamting the cell cycle time-dependent RNA kinetic rates with the test data provided:

Load required libraries

Install the required library if it is not installed yet.

library(tibble)
library(dplyr)
library(magrittr)
library(tidyr)
library(ggplot2)
library(cowplot)
library(Rmpfr)
library(gam)
library(mgcv)
library(foreach)
library(doParallel)
library(Eskrate)

Input data

  1. cell meta data: cell_info containing columns of cell_id and cc_time.

  2. raw gene expression data: matrices of raw gene expression data of four RNA types containing columns of cell_id, gene, capture_rate, p_u, p_total, m_u, m_total.

### cell meta data
cells_info <- readRDS("/data/cells_info.rds"))

### gene expression matrix
raw_labeled_mature <- readRDS("/data/raw_counts_labeled_mature.rds"))
raw_unlabeled_mature <- readRDS("/data/raw_counts_unlabeled_mature.rds"))
raw_labeled_precursor <- readRDS( "/data/raw_counts_labeled_precursor.rds"))
raw_unlabeled_precursor <- readRDS("/data/raw_counts_unlabeled_precursor.rds"))

observed.data <- 
  getInput(rawCountsLabeledMature = raw_labeled_mature,
           rawCountsUnlabeledMature = raw_unlabeled_mature,
           rawCountsLabeledPrecursor = raw_labeled_precursor,
           rawCountsUnlabeledPrecursor = raw_unlabeled_precursor,
           cells = cells_info,
           labelingTime = 15)

### select two genes (ZNF711, KIF14) for demo
observed.data %<>% 
  filter(gene %in% c("ZNF711", "KIF14")) %>%
  droplevels

Estimate RNA kinetic rates of transcription and degradation

data_rates <- 
  estimateKinetics(observedData = observed.data, 
                   threadN = 2)

Calculate predictions of gene expression using the estimated kinetic rates

data_predicted <- 
  getPredictions(dataRates = data_rates, 
                 threadN=2)

Plot predicted gene expression and estimated rates

For example, plot PCNA gene expression and rates.

p1 <-
  plotMtotalAlphaGamma(df.prediction = predicted.data,   
                       gene.to.plot = "ZNF711",
                       color.alpha = "#D95F02",
                       color.gamma = "#1B9E77",
                       color.expression = "#666666",
                       show.phase.boundaries = FALSE,
                       line.size = 1,
                       font.size = 16)

p2 <-
  plotMtotalAlphaGamma(df.prediction = predicted.data,   
                       gene.to.plot = "KIF14",
                       color.alpha = "#D95F02",
                       color.gamma = "#1B9E77",
                       color.expression = "#666666",
                       show.phase.boundaries = FALSE,
                       line.size = 1,
                       font.size = 16)

plot_grid(p1, p2, ncol=2)

Reference

About

No description, website, or topics provided.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages