sashaphanes’s gists

sashaphanes / ggplot2BoxplotGrouping.R

Last active December 17, 2015 04:29

	#A solution with ggplot2

	#library(reshape2)
	#test.m <- melt(test)

	#sample of dataframe:
	head(combined.data)
	> region expression species
	1 NAc 0.4711284 macaque
	2 NAc -0.1003201 macaque

sashaphanes / combinedANOVA.R

Created May 9, 2013 21:09

	#read combined data
	setwd("/home/rosseraa/work.dir/allen.brain/")
	raw.data = read.csv("common.regions.ANOVA.csv")

	#run ANOVA
	aov.result = aov(expression ~ region * species, data=raw.data)
	summary(aov.result)
	TukeyHSD(aov.result)
	summary(aov.result)[[1]][["Pr(>F)"]]
	print(model.tables(aov.result, "means"), digits=4)

sashaphanes / sidebysideANOVAboxplots.R

Created May 9, 2013 21:08

	#read macaque data
	setwd("/home/rosseraa/work.dir/allen.brain/")
	macaque.raw.data = read.csv("macaque.common.regions.ANOVA.csv")

	#standardize (z-score) macaque expression values
	macaque.raw.data$expression = scale(macaque.raw.data$expression)

	#run macaque ANOVA
	aov.result = aov(expression ~ region, data=macaque.raw.data)
	summary(aov.result) #is there a significant difference between any of the group means? If so, move onto a post-hoc test

sashaphanes / human.maoa.ANOVA.R

Last active December 17, 2015 02:19

	#read human data
	setwd("/home/rosseraa/work.dir/allen.brain/")
	human.raw.data = read.csv("human.common.regions.ANOVA.csv")

	#run human ANOVA
	aov.result = aov(expression ~ region, data=human.raw.data)
	summary(aov.result) #is there a significant difference between any of the group means? If so, move onto a post-hoc test
	TukeyHSD(aov.result) #post-hoc for multiple pairwise comparisons to see where the significant differences lie.
	summary(aov.result)[[1]][["Pr(>F)"]] #p-values
	print(model.tables(aov.result, "means"), digits=4)

sashaphanes / macaque.ANOVA.R

Last active December 17, 2015 01:09

	#read macaque data
	setwd("/home/rosseraa/work.dir/allen.brain/")
	raw.data = read.csv("macaque.common.regions.ANOVA.csv")

	#standardize expression values
	raw.data$expression = scale(raw.data$expression)

	#run ANOVA
	aov.result = aov(expression ~ region, data=raw.data)
	summary(aov.result) #is there a significant difference between any of the group means? If so, move onto a post-hoc test

sashaphanes / macaque.analysis.R

Last active December 16, 2015 20:59

	#Thsdf
	#read macaque data
	setwd("/home/rosseraa/work.dir/allen.brain/macaque")
	raw.data = read.csv("maoaOut.csv")

	#get pairwise combinations
	com = combn(colnames(raw.data), 2)

	#get p-values
	p.values = apply(com, 2, function(x) t.test(raw.data[,x[1]], raw.data[,x[2]])$p.val)

sashaphanes / simple_web_form.php

Last active December 16, 2015 16:59

web form

	#index.php
	<html>
	<form name="input" action="html_form_action.php" method="get">
	Please enter your name: <input type="text" name="user_name">
	<input type="submit" value="Submit">
	</form>
	</html>

sashaphanes / homozygousAltImpact.bash

Created April 19, 2013 18:55

Sanger recently released mouse SNPs (in VCF format) from next-generation sequencing for 18 strains. Parse all SNPs where the alternative alleles are homozygous for strain 129S1 for gene Impact, no local save #ftp://ftp-mouse.sanger.ac.uk/current_snps/mgp.v3.snps.rsIDdbSNPv137.vcf.gz



	lftp -c 'open -e "zcat mgp.v3.snps.rsIDdbSNPv137.vcf.gz” ftp-mouse.sanger.ac.uk/current_snps/' \| perl -ne '{chomp; if (/^#CHROM/) {print "$_\n"; }else {@a = split (/\t/, $_); print "$_\n" if ($a[0] ==18 and $a[1] >= 12972252 and $a[1] <= 12992948 and $a[10] =~ /^1\/1\|^1\\|1/); } }' > mm10.129S1.Impact.altHom &

sashaphanes / qnormdata.R

Created April 19, 2013 18:42

The following is a real data file and contains four different types of gene expression measurements (including RPKM) from one most recent finished RNAseq data. Find the file path from your Linux machine or rhdna account and load the data directly into R. Then select RPKM measurements for all samples. After excluding genes that no samples have RP…

	http://156.40.185.32/pub/human.Hela.ALDH.ethanol.RNAseq/features/allHela.sorted.wig.exon.high50Ave.peakFeats.allChr.exonLen.subTotal.RPKM.highest_in_gene.transposed.txt.gz
	README file: http://156.40.185.32/pub/human.Hela.ALDH.ethanol.RNAseq/README.human.Hela.ALDH.ethanol.RNAseq.txt

	dataFile = "/mnt/xdat/pub/human.Hela.ALDH.ethanol.RNAseq/features/allHela.sorted.wig.exon.high50Ave.peakFeats.allChr.exonLen.subTotal.RPKM.highest_in_gene.transposed.txt.gz"
	rawdata = read.delim(gzfile(dataFile), sep="\t", header=T)
	rownames(rawdata) = paste(rawdata[,"Chr"], rawdata[,"GeneSym"], sep = ".")
	data.RPKM = rawdata[, grep("RPKM", colnames(rawdata))]
	rowMax = do.call("pmax", c(data.RPKM[,1:ncol(data.RPKM)],na.rm=TRUE))
	data.RPKM.rowMax.gt1 = data.RPKM[rowMax >1,]
	data.RPKM.rowMax.gt1[is.na(data.RPKM.rowMax.gt1)] = 0

sashaphanes / binom.R

Created April 19, 2013 18:41

RNAseq reads covered 5,000 biallelic human autosomal SNPs (all are heterogeneous, assume they are not in LD). Simple R statements plot the distribution of the alternative allelic frequency under sequencing coverage 2x, 5x, 10x, 20x, respectively. Calculate the p-values with the hypothesis that there are no differential allelic expression for the…

	plot(x=c(0:2)/2, dbinom(c(0:2),2,0.5), pch=20, col="red", type ="b", xlim =c(0,1), ylim =c(0, 0.6), xlab = "AltFreq", ylab="Probability" )
	lines (x=c(0:5)/5, dbinom(c(0:5),5,0.5), pch=20, col="brown",type ="b" )
	lines (x=c(0:10)/10, dbinom(c(0:10),10,0.5), pch=20, col="green",type ="b")
	lines (x=c(0:20)/20, dbinom(c(0:20),20,0.5), pch=20, col="blue",type ="b")

	legend(0.8, 0.58, legend="2x", col="red", pch=20, lty = 1, bty="n" )
	legend(0.8, 0.56, legend="5x", col="brown", pch=20, lty = 1, bty="n" )
	legend(0.8, 0.54, legend="10x", col="green", pch=20, lty = 1, bty="n" )
	legend(0.8, 0.52, legend="20x", col="blue", pch=20, lty = 1, bty="n" )