mikelove

library(plyranges)
x <- data.frame(seqnames=1,
                start=sample(1000,10,FALSE),
                width=1, id=1:10) %>%
  as_granges()
y <- data.frame(seqnames=1,
                start=sample(1000,10,FALSE),
                width=1, id=letters[1:10]) %>%
  as_granges()

mikelove

library(plyranges)
library(microbenchmark)
library(dplyr)
library(tibble)

make_rand_gr <- function(N, grps) {
  data.frame(seqnames = sample(c("seq1", "seq2", "seq3"), N, replace = TRUE), 
    strand = sample(c("+", "-", "*"), N, replace = TRUE), start = rpois(N, 
      N), width = rpois(N, N), grps = sample(grps, N, replace = TRUE), 
    score = runif(N)) %>% as_granges()

mikelove

airway@assays@data[["counts"]][airway@rowRanges@elementMetadata@listData$gene_id == "ENSG00000000003", airway@colData@listData$dex == "trt"]
assays(airway)[["counts"]][names(rowRanges(airway)) == "ENSG00000000003", colData(airway)$dex == "trt"]
assay(airway, "counts")["ENSG00000000003", airway$dex == "trt"]
airway |> filter(symbol == "TSPAN6", dex == "trt") |> pull(counts)

mikelove

# scraped from https://doi.org/10.1038/nature11241
# note that intersections < 100 not included
banana_sets <- c(
	Musa = 759,
	Phoenix = 769,
	Sorghum = 827,
	Brachypodium = 387,
	Ozyza = 1246,
	Arabidopsis = 1187,
	`Musa&Phoenix` = 467,

mikelove

library(plyranges)

set.seed(1)
x <- data.frame(seqnames=1, start=0:9 * 100 + 1, 
                width=20, id=1:10) %>%
  as_granges()

y <- data.frame(seqnames=1, start=round(runif(4,100,900)), 
                width=10, id=letters[1:4]) %>%
  as_granges() %>%

mikelove

library(splines)
library(DESeq2)

# make some demo data
dds <- makeExampleDESeqDataSet(n=100, m=40)
dds$condition <- sort(runif(40))
# make one gene where expression has a curve shape (just for demo)
s_shape <- round(500 * sin(dds$condition*2*pi) + 1000 + rnorm(40,0,50))
mode(s_shape) <- "integer"
counts(dds)[1,] <- s_shape

mikelove

library(tm)
library(wordcloud)
crude <- scan("words", what="char", sep="\n")
crude <- gsub("/"," ",crude)
crude <- gsub("single cell","singlecell",crude)
crude <- Corpus(VectorSource(crude))
crude <- suppressWarnings(tm_map(crude, removePunctuation))
crude <- suppressWarnings(tm_map(crude, function(x) removeWords(x, stopwords())))
tdm <- TermDocumentMatrix(crude)
m <- as.matrix(tdm)

mikelove

design <- ~0 + donor + allele
dds <- DESeqDataSetFromMatrix(counts, coldata, design) # counts has the two alleles per donor
assays(dds)[["weights"]] <- weights # 1 for hets, 1e-6 for homs
sizeFactors(dds) <- rep(1, ncol(dds))
dds <- DESeq(dds, test="LRT", reduced=~0 + donor)
res <- results(dds)

mikelove

library(plotgardener)
par <- pgParams(
  chrom = "chr1", 
  chromstart = 8.2e6,
  chromend = 8.7e6,
  assembly = "hg19",
  just = c("left", "bottom")
)
dat <- data.frame(chrom="chr1", pos=840:849 * 1e4 + 1, p=1:10/10)
pageCreate(width = 4, height = 3.5, showGuides = TRUE)

mikelove

library(dplyr)
library(ggplot2)
library(tidyr)
library(purrr)
library(broom)
library(forcats)

d <- data.frame(
  value = rnorm(24),
  type = factor(rep(c("A","B"), c(5,3))),