mikelove

---
title: "Differential chromatin accessibility and gene expression"
format: html
---

# Differential expression from RNA-seq

```{r}
#| eval: FALSE
dir <- system.file("extdata", package="macrophage")

mikelove

mikelove

library(plyranges)
library(nullranges)

x <- data.frame(
  seqnames=rep(c("1","2","3"), each=10),
  start=rep(c(0:4, 10:14) * 1000 + 1, times=3),
  width=100) |>
  as_granges()

seqlengths(x) <- c("1"=20123, "2"=20123, "3"=20123)

mikelove

library(SummarizedExperiment)
library(plyranges)

# example data
m <- matrix(rnorm(600), nrow=100)
r1 <- data.frame(seqnames=1, start=1:50 * 100 + 2501,
                 width=5, id1=paste0("u",formatC(1:50,width=3,flag="0"))) |>
  as_granges()
r2 <- data.frame(seqnames=1, start=1:100 * 100 + 1,
                 width=5, id2=paste0("v",formatC(1:100,width=3,flag="0"))) |>

mikelove

library(shiny)
library(UpSetR)
library(dplyr)
library(tidyr)
library(readr)
library(ggplot2)
dat <- read_delim("ancestry_dataframe.tsv")

ui <- fluidPage(
    titlePanel("IGVF Ancestry Dashboard"),

mikelove

mat <- matrix(rnbinom(2e5, mu=100, size=1/.01), ncol=100)

library(DESeq2)

d <- DESeqDataSetFromMatrix(mat, DataFrame(x=rep(1,100)), ~1)
# library size correction, centered log ratio to reference sample
d <- estimateSizeFactors(d)
# variance
d <- estimateDispersionsGeneEst(d)
# trend

mikelove

set.seed(5)

n <- 1000
reps <- 10
rna <- matrix(
  rnbinom(n * reps, mu = 10, size = 100),
  ncol=reps
)
dna <- matrix(
  rnbinom(n * reps, mu = 10, size = 100),

mikelove

# dataframes vs lm S3 vs Bioc S4
# Michael Love
# Nov 1 2023

dat <- data.frame(genotype=c("wt","wt","mut","mut"),
                  count=c(10,20,30,40),
                  score=c(-1.2,0,3.4,-5),
                  gene=c("Abc","Abc","Xyz","Xyz"))
library(tibble)
dat |> as_tibble()

mikelove

---
title: "Toy tree example for collapsing"
author: "Michael Love"
---
  
Example data with 20 inferential replicates, here we just have 1
sample per condition and we calculate the LFC at each level of the
tree.

From the below simulation setup (see first chunk), the true DE signal

mikelove

library(plyranges)
library(readr)

bindata <- read_tsv("bindata.40000.hg19.tsv.gz")

fire <- read_bed("fire-adult-hg19.txt")

# not necessary, but nice to have
si <- Seqinfo(genome="hg19")
si <- keepStandardChromosomes(si)