help()
dim(state.x77)
state.x77[1:10,]
state
state.region
aggregate(presidents, nf = 1, FUN = mean)
y <- c(1:3,7,5)#
x <- c(1:3,6:7)#
( ee <- effects(lm(y ~ x)) )#
c(round(ee - effects(lm(y+10 ~ I(x-3.8))),3))# just the first is different
ee
effects(lm(y+10 ~ I(x-3.8)))
I(x-3.8)
(x-3.8)
class(x-3.8)
class(I(x-3.8))
lm(y ~ x)
x <- 1:5; coef(lm(c(1:3,7,6) ~ x))
memory.profile()
 format(rep(13.7,3), width=10,nsmall=3)
ls()
tj <- read.csv(stdin(),comment.char="#")
tg<- tj
tg
td <- tj[,3:ncol(tj)]
td
rownames(td)<-tj[,1]
td
h <- hist(td)
dim(td)
VADeaths
barplot(td)
class9td)
class(td)
barplot(as.matrix(td))
barplot(as.matrix(td),beside=T)
barplot(as.matrix(td["Daphx_GNO"]),beside=T)
barplot(as.matrix(td["Daphx_GNO",]),beside=T)
barplot(as.matrix(td["Daphx_GNO",-1]),beside=T)
barplot(as.matrix(td[,-1]),beside=T)
td[,-1]
td[,-1]/td[,1]
barplot(as.matrix(td[,-1]/td[,1]),beside=T)
setwd("~/Desktop/dspp-work/genomesoft/tandy/")
load(".RData")
ls()
gpa
tw1
pe1 <- read.table("pexons.tandynear4.txt", header=T, strip.white=T)
dim(pe1)
pe1
pe1$species
pe1$group
pe1$ggroup
pe <- pe1
pelev <- grep("NCBI",as.character(pe$ggroup),value=T)
pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
pelev
pefac
pea <-  pe[!is.na(pefac),c(-1,-2)]
paste( pe[!is.na(pefac),c(1,2)], sep="")
pea
pe[!is.na(pefac),c(1,2)]
paste( pe[!is.na(pefac),c(1,2)])
paste( as.character(pe[!is.na(pefac),c(1,2)]))
pe[!is.na(pefac),1]
pe[!is.na(pefac),2]
factors(pe[!is.na(pefac),2])
factor(pe[!is.na(pefac),2])
factor(pe[!is.na(pefac),c(1,2)])
colnames(pea)
rownames(pea) <- pe[!is.na(pefac),1] # species
pea
pelev
maxgrep("NCBI",as.character(pe$ggroup),value=T)
grep("NCBI",as.character(pe$ggroup),value=T)
grep("_found",as.character(colnames(pea)))
predtitle <- "Pred. exons/species, predictor="
ylabel <- "exon count"
  tparm <- grep("_found",as.character(colnames(pea))) #   2  4  6  8 10 12
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),
grep("_found",as.character(colnames(pea)))
grep("_found",as.character(colnames(pea)),value=T)
grep("!Same",grep("_found",as.character(colnames(pea)),value=T),value=T)
grep("_found",colnames(pea)) - grep("Same",colnames(pea))
grep("_found",colnames(pea)) & grep("Same",colnames(pea))
grep("Same",colnames(pea))
grep("_found",colnames(pea))
grep("_found",colnames(pea)) ! grep("Same",colnames(pea))
grep("_found",colnames(pea)) %!% grep("Same",colnames(pea))
grep("_found",colnames(pea)) %*% grep("Same",colnames(pea))
regexpr("_found",colnames(pea))
gregexpr("_found",colnames(pea))
regexpr("_found",colnames(pea))
regexpr("_found",as.character(colnames(pea)))
colnames(pea)
regexpr("_found",colnames(pea))
regexpr("Same",colnames(pea))
-regexpr("Same",colnames(pea))
-regexpr("Same",colnames(pea))[1]
-regexpr("Same",colnames(pea))[1,]
-regexpr("Same",colnames(pea))[,1]
- regexpr("Same",colnames(pea))
pmatch("Same",colnames(pea))
pmatch("_found",colnames(pea))
pmatch("found",colnames(pea))
pmatch("found",colnames(pea),duplicates.ok=T)
grep("_found",colnames(pea))
grep("Same",colnames(pea))
colnames(pea)[1]
colnames(pea)["align"]
grep("align",colnames(pea))
grep("nxfound",colnames(pea))
grep("found",colnames(pea))
grep("^found",colnames(pea))
  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  # tparm <- grep("_found",as.character(colnames(pea))) #   2  4  6  8 10 12
  tparm <- c(2,4,6,8,10,15) # 13,17 == align,nxfound
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),
ggroup <- "Gnomon"
predtitle <- paste("Pred. exons/species, predictor",ggroup)
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),

  legend( x="topright", colnames(pea[,tparm]), lwd=3, col=tcols)
sppset <- c("dmel","dsec","dsim","dyak","dere","dana","dpse","dper",
          "dwil","dmoj","dvir","dgri","dappulx","caeele")
spat <- paste(sppset,collapse="|")
rown <- pe[!is.na(pefac),1] # species
rown
rown[sppset,]
rown
sppset1 <- c("dsec","dyak","dere","dpse","dwil","dgri","dappulx")
rown[sppset1,]
dim(sppset1)
length(sppset1)
dim(rown)
length(rown)
order(rown,sppset1)
order(sppset1,rown)
rown[ order(sppset1,rown) ]
pea[ order(sppset1,rown), ]
pea[ order(sppset1), ]
pea[ sppset1, ]
pea <-  pe[!is.na(pefac),c(-1,-2)]
rown <- pe[!is.na(pefac),1] # species
rownames(pea) <- rown
pea <- pea[sppset1,]
pea
  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  tparm <- c( 6,8,10,15) # 13,17 == align,nxfound   ; 2,4, = far, inside
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),

  legend( x="topright", colnames(pea[,tparm]), lwd=3, col=tcols)
pea
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
pea[,tparm] / pea$Same_found
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),

  legend( x="topright", colnames(pea[,tparm]), lwd=3, col=tcols)
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
pe
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI",as.character(pe$ggroup),value=T)
  sppset <- c("dsec","dyak","dere","dpse","dwil","dgri","dappulx")
  spat <- paste(sppset,collapse="|")
  
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]
pea
  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"
  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    beside=T, legend=F, col = tcols ) #  ylim=c(0,ymax),

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
  ymax <- 0.05 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
 tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ggroup <- "Gnomon"
pelev <- grep("NCBI",as.character(pe$ggroup),value=T)
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"

  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  ymax <- 0.05 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
pea
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI",as.character(pe$ggroup),value=T)
pelev
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
pefac
  pea <-  pe[!is.na(pefac),c(-1,-2)]
pea
  pea <-  pe[!is.na(pefac),]
pea
  rown <- pe[!is.na(pefac),1] # species
rown
  rownames(pea) <- rown
pea
pea[sppset,]
pe <- read.table("pexons.tandynear4.txt", header=T, strip.white=T)
ggroup <- "Gnomon"
pelev <- grep("NCBI",as.character(pe$ggroup),value=T)
  sppset <- c("dsec","dyak","dere","dana","dpse","dwil","dgri","dappulx")
  spat <- paste(sppset,collapse="|")
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"
  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  ymax <- 0.05 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  ymax <- 0.04 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
pe <- read.table("pexons.tandynear4.txt", header=T, strip.white=T)
pe
pe <- read.table("pexons.tandynear4.txt", header=T, strip.white=T)
pe
ggroup <- "Gnomon"
pelev <- grep("NCBI",as.character(pe$ggroup),value=T) 
# should paste species, ggroup as one factor
  sppset <- c("dsec","dyak","dere","dana","dpse","dwil","dgri","dappulx")
  spat <- paste(sppset,collapse="|")
  
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"
  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  ymax <- 0.04 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
pea
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI",as.character(pe$ggroup),value=T)
pea.gno <- pea
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"

  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  tparm <- c( 6,8,10) # ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  ymax <- 0.04 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  
  
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
pea.glr <- pea
pea.gno
pea.glr
pe <- read.table("pexons.tandynear4.txt", header=T, strip.white=T)
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI",as.character(pe$ggroup),value=T)
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI|FBtr",as.character(pe$ggroup),value=T)
pelev
  sppset <- c("dmel","dsec","dyak","dere","dana","dpse","dwil","dgri","dappulx")
  spat <- paste(sppset,collapse="|")
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"
pea
  tparm <- c( 6,8,10) # near_found ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  tparm <- c( 5,7,9) # near ; lacks dyak odd spike > dyak tandems were found at higher rate than others?
  
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  ymax <- 0.04 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ymax <- 0.20
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ymax <- 0.15
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ymax <- 0.20
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ggroup <- "Gnomon"
pelev <- grep("NCBI",as.character(pe$ggroup),value=T) 
# should paste species, ggroup as one factor
 ymax <- 0.15
ggroup <- "GleanR"
pelev <- grep("GLEANR|JGI|FBtr",as.character(pe$ggroup),value=T)
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ggroup <- "Gnomon"
pelev <- grep("NCBI",as.character(pe$ggroup),value=T) 
# should paste species, ggroup as one factor
  sppset <- c("dmel","dsec","dyak","dere","dana","dpse","dwil","dgri","dappulx")
  spat <- paste(sppset,collapse="|")
  
  pefac <- factor(pe$ggroup, levels=pelev, ordered=T,exclude=NULL)
  
  pea <-  pe[!is.na(pefac),c(-1,-2)]
  rown <- pe[!is.na(pefac),1] # species
  rownames(pea) <- rown
  pea <- pea[sppset,]

  predtitle <- paste("Pred. exons/species, predictor",ggroup)
  ylabel <- "exon count"

  pea$align <- pea$align/pea$cdslen
  pea$nxfound <- pea$nxfound/pea$nexon
  
  tparm <- c( 6,8,10) # near_found ,15 = found ~= nearcount; 13,17 == align,nxfound   ; 2,4, = far, inside
  ymax <- 0.04 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR

  tparm <- c( 5,7,9) # near ; lacks dyak odd spike > dyak tandems were found at higher rate than others?
  ymax <- 0.15 # w/ above as relative Same, max is 0.03+ for Gnomon, 0.015 for GleanR
  
  pea[,tparm] <- pea[,tparm] / pea$Same_found
  tcols <- c("lightblue", "mistyrose", "lightcyan", "lavender", 2:20) [1:length(tparm)]
  
  bpx <- barplot( as.matrix(t( pea[,tparm])),ylab=ylabel, main=predtitle,
    ylim=c(0,ymax), beside=T, legend=F, col = tcols ) #  

  legend( x="topleft", colnames(pea[,tparm]), lwd=3, col=tcols)
ls()