Hello everyone,
The following is a portion of coding that a colleague sent. Given my lack
of experience in R, I am not quite sure what the significance of the
following arguments. Could anyone help me translate? For context, I am
aware of the downloading portion of the script... library(data.table) etc.,
but am not familiar with the portion pertaining to an1 .
library(data.table)
anno = as.data.frame(fread(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/450K/mapper.txt", sep ="\t",
header = T))
meth = read.table(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/27K/GBM.txt", sep ="\t",
header = T, row.names = 1)
meth = as.matrix(meth)
""" the loop just formats the methylation column names to match format"""
colnames(meth) = sapply(colnames(meth), function(i){
c1 = strsplit(i,split = '.', fixed = T)[[1]]
c1[4] = paste(strsplit(c1[4],split = "",fixed = T)[[1]][1:2],collapse =
"")
paste(c1,collapse = ".")
})
exp = read.table(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/RNAseq/GBM.txt", sep = "\t",
header = T, row.names = 1)
exp = as.matrix(exp)
c = intersect(colnames(exp),colnames(meth))
exp = exp[,c]
meth = meth[,c]
m = apply(meth, 1, function(i){
log2(i/(1-i))
})
m = t(as.matrix(m))
an = anno[anno$probe %in% rownames(m),]
an = an[an$gene %in% rownames(exp),]
an = an[an$location %in% c("TSS200","TSS1500"),]
p = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~
m[as.character(i[1]),]))$coefficient[2,4], error= function(e)NA)
})
t = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~
m[as.character(i[1]),]))$coefficient[2,3], error= function(e)NA)
})
an1 =cbind(an,p)
an1 = cbind(an1,t)
an1$q = p.adjust(as.numeric(an1$p))
summary(lm(exp["MAOB",] ~ m["cg00121904",]$coefficient[2,c(3:4)]
###############################################
R Data
4 messages · Spencer Brackett, Fowler, Mark
Hi Spencer,
The an1 syntax is adding regression coefficients (or NAs where a regression could not be done) to the downloaded and processed data, which ends up a matrix. The cbind function adds the regression coefficients to the last column of the matrix (i.e. bind the columns of the inputs in the order given). Simple example below. Not actually any need for the separate cbind commands, could have just used an1=cbind(an,p,t). The cbind function expects all the columns to be of the same length, hence the use of the tryCatch function to capture NA's for failed regression attempts, ensuring that p and t correspond row by row with the matrix.
x=seq(1,5)
y=seq(6,10)
z=seq(1,5)
xyz=cbind(x,y,z)
xyz
x y z
[1,] 1 6 1
[2,] 2 7 2
[3,] 3 8 3
[4,] 4 9 4
[5,] 5 10 5
dangs=rep(NA,5)
xyzd=cbind(xyz,dangs)
xyzd
x y z dangs
[1,] 1 6 1 NA
[2,] 2 7 2 NA
[3,] 3 8 3 NA
[4,] 4 9 4 NA
[5,] 5 10 5 NA
-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Spencer Brackett
Sent: February 14, 2019 12:32 AM
To: R-help <r-help at r-project.org>; Sarah Goslee <sarah.goslee at gmail.com>; Caitlin Gibbons <bioprogrammer at gmail.com>; Jeff Newmiller <jdnewmil at dcn.davis.ca.us>
Subject: [R] R Data
Hello everyone,
The following is a portion of coding that a colleague sent. Given my lack of experience in R, I am not quite sure what the significance of the following arguments. Could anyone help me translate? For context, I am aware of the downloading portion of the script... library(data.table) etc., but am not familiar with the portion pertaining to an1 .
library(data.table)
anno = as.data.frame(fread(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/450K/mapper.txt", sep ="\t", header = T)) meth = read.table(file = "/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/27K/GBM.txt", sep ="\t", header = T, row.names = 1) meth = as.matrix(meth) """ the loop just formats the methylation column names to match format"""
colnames(meth) = sapply(colnames(meth), function(i){
c1 = strsplit(i,split = '.', fixed = T)[[1]]
c1[4] = paste(strsplit(c1[4],split = "",fixed = T)[[1]][1:2],collapse =
"")
paste(c1,collapse = ".")
})
exp = read.table(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/RNAseq/GBM.txt", sep = "\t", header = T, row.names = 1) exp = as.matrix(exp) c = intersect(colnames(exp),colnames(meth))
exp = exp[,c]
meth = meth[,c]
m = apply(meth, 1, function(i){
log2(i/(1-i))
})
m = t(as.matrix(m))
an = anno[anno$probe %in% rownames(m),]
an = an[an$gene %in% rownames(exp),]
an = an[an$location %in% c("TSS200","TSS1500"),]
p = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,4], error= function(e)NA)
})
t = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,3], error= function(e)NA)
})
an1 =cbind(an,p)
an1 = cbind(an1,t)
an1$q = p.adjust(as.numeric(an1$p))
summary(lm(exp["MAOB",] ~ m["cg00121904",]$coefficient[2,c(3:4)]
###############################################
______________________________________________
R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.
Mr. Fowler, Thank you! This information is most helpful. So from my understanding, I can use the regression coefficients shown (via the coding I originally sent, to generate a continuous distribution with what is essentially a line of best fit? The data added here had some 30,000 variables (it is genomic data from TCGA), does this mean that any none NA data is being accounted for in said distribution? Best, Spencer Brackett On Thursday, February 14, 2019, Fowler, Mark <Mark.Fowler at dfo-mpo.gc.ca> wrote:
Hi Spencer,
The an1 syntax is adding regression coefficients (or NAs where a
regression could not be done) to the downloaded and processed data, which
ends up a matrix. The cbind function adds the regression coefficients to
the last column of the matrix (i.e. bind the columns of the inputs in the
order given). Simple example below. Not actually any need for the separate
cbind commands, could have just used an1=cbind(an,p,t). The cbind function
expects all the columns to be of the same length, hence the use of the
tryCatch function to capture NA's for failed regression attempts, ensuring
that p and t correspond row by row with the matrix.
x=seq(1,5)
y=seq(6,10)
z=seq(1,5)
xyz=cbind(x,y,z)
xyz
x y z
[1,] 1 6 1
[2,] 2 7 2
[3,] 3 8 3
[4,] 4 9 4
[5,] 5 10 5
dangs=rep(NA,5)
xyzd=cbind(xyz,dangs)
xyzd
x y z dangs
[1,] 1 6 1 NA
[2,] 2 7 2 NA
[3,] 3 8 3 NA
[4,] 4 9 4 NA
[5,] 5 10 5 NA
-----Original Message-----
From: R-help <r-help-bounces at r-project.org> On Behalf Of Spencer Brackett
Sent: February 14, 2019 12:32 AM
To: R-help <r-help at r-project.org>; Sarah Goslee <sarah.goslee at gmail.com>;
Caitlin Gibbons <bioprogrammer at gmail.com>; Jeff Newmiller <
jdnewmil at dcn.davis.ca.us>
Subject: [R] R Data
Hello everyone,
The following is a portion of coding that a colleague sent. Given my lack
of experience in R, I am not quite sure what the significance of the
following arguments. Could anyone help me translate? For context, I am
aware of the downloading portion of the script... library(data.table) etc.,
but am not familiar with the portion pertaining to an1 .
library(data.table)
anno = as.data.frame(fread(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/450K/mapper.txt", sep ="\t",
header = T)) meth = read.table(file = "/rsrch1/bcb/kchen_group/v_
mohanty/data/TCGA/27K/GBM.txt", sep ="\t", header = T, row.names = 1)
meth = as.matrix(meth) """ the loop just formats the methylation column
names to match format"""
colnames(meth) = sapply(colnames(meth), function(i){
c1 = strsplit(i,split = '.', fixed = T)[[1]]
c1[4] = paste(strsplit(c1[4],split = "",fixed = T)[[1]][1:2],collapse =
"")
paste(c1,collapse = ".")
})
exp = read.table(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/RNAseq/GBM.txt", sep = "\t",
header = T, row.names = 1) exp = as.matrix(exp) c = intersect(colnames(exp),
colnames(meth))
exp = exp[,c]
meth = meth[,c]
m = apply(meth, 1, function(i){
log2(i/(1-i))
})
m = t(as.matrix(m))
an = anno[anno$probe %in% rownames(m),]
an = an[an$gene %in% rownames(exp),]
an = an[an$location %in% c("TSS200","TSS1500"),]
p = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,4],
error= function(e)NA)
})
t = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,3],
error= function(e)NA)
})
an1 =cbind(an,p)
an1 = cbind(an1,t)
an1$q = p.adjust(as.numeric(an1$p))
summary(lm(exp["MAOB",] ~ m["cg00121904",]$coefficient[2,c(3:4)]
###############################################
[[alternative HTML version deleted]]
______________________________________________ R-help at r-project.org mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help PLEASE do read the posting guide http://www.R-project.org/ posting-guide.html and provide commented, minimal, self-contained, reproducible code.
I am not sure I would use the word ?accounted?, more like discounted (tossed out). From: Spencer Brackett <spbrackett20 at saintjosephhs.com> Sent: February 14, 2019 9:21 AM To: Fowler, Mark <Mark.Fowler at dfo-mpo.gc.ca> Cc: R-help <r-help at r-project.org>; Sarah Goslee <sarah.goslee at gmail.com>; Caitlin Gibbons <bioprogrammer at gmail.com>; Jeff Newmiller <jdnewmil at dcn.davis.ca.us> Subject: Re: R Data Mr. Fowler, Thank you! This information is most helpful. So from my understanding, I can use the regression coefficients shown (via the coding I originally sent, to generate a continuous distribution with what is essentially a line of best fit? The data added here had some 30,000 variables (it is genomic data from TCGA), does this mean that any none NA data is being accounted for in said distribution? Best, Spencer Brackett
On Thursday, February 14, 2019, Fowler, Mark <Mark.Fowler at dfo-mpo.gc.ca<mailto:Mark.Fowler at dfo-mpo.gc.ca>> wrote:
Hi Spencer,
The an1 syntax is adding regression coefficients (or NAs where a regression could not be done) to the downloaded and processed data, which ends up a matrix. The cbind function adds the regression coefficients to the last column of the matrix (i.e. bind the columns of the inputs in the order given). Simple example below. Not actually any need for the separate cbind commands, could have just used an1=cbind(an,p,t). The cbind function expects all the columns to be of the same length, hence the use of the tryCatch function to capture NA's for failed regression attempts, ensuring that p and t correspond row by row with the matrix.
x=seq(1,5)
y=seq(6,10)
z=seq(1,5)
xyz=cbind(x,y,z)
xyz
x y z
[1,] 1 6 1
[2,] 2 7 2
[3,] 3 8 3
[4,] 4 9 4
[5,] 5 10 5
dangs=rep(NA,5)
xyzd=cbind(xyz,dangs)
xyzd
x y z dangs
[1,] 1 6 1 NA
[2,] 2 7 2 NA
[3,] 3 8 3 NA
[4,] 4 9 4 NA
[5,] 5 10 5 NA
-----Original Message-----
From: R-help <r-help-bounces at r-project.org<mailto:r-help-bounces at r-project.org>> On Behalf Of Spencer Brackett
Sent: February 14, 2019 12:32 AM
To: R-help <r-help at r-project.org<mailto:r-help at r-project.org>>; Sarah Goslee <sarah.goslee at gmail.com<mailto:sarah.goslee at gmail.com>>; Caitlin Gibbons <bioprogrammer at gmail.com<mailto:bioprogrammer at gmail.com>>; Jeff Newmiller <jdnewmil at dcn.davis.ca.us<mailto:jdnewmil at dcn.davis.ca.us>>
Subject: [R] R Data
Hello everyone,
The following is a portion of coding that a colleague sent. Given my lack of experience in R, I am not quite sure what the significance of the following arguments. Could anyone help me translate? For context, I am aware of the downloading portion of the script... library(data.table) etc., but am not familiar with the portion pertaining to an1 .
library(data.table)
anno = as.data.frame(fread(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/450K/mapper.txt", sep ="\t", header = T)) meth = read.table(file = "/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/27K/GBM.txt", sep ="\t", header = T, row.names = 1) meth = as.matrix(meth) """ the loop just formats the methylation column names to match format"""
colnames(meth) = sapply(colnames(meth), function(i){
c1 = strsplit(i,split = '.', fixed = T)[[1]]
c1[4] = paste(strsplit(c1[4],split = "",fixed = T)[[1]][1:2],collapse =
"")
paste(c1,collapse = ".")
})
exp = read.table(file =
"/rsrch1/bcb/kchen_group/v_mohanty/data/TCGA/RNAseq/GBM.txt", sep = "\t", header = T, row.names = 1) exp = as.matrix(exp) c = intersect(colnames(exp),colnames(meth))
exp = exp[,c]
meth = meth[,c]
m = apply(meth, 1, function(i){
log2(i/(1-i))
})
m = t(as.matrix(m))
an = anno[anno$probe %in% rownames(m),]
an = an[an$gene %in% rownames(exp),]
an = an[an$location %in% c("TSS200","TSS1500"),]
p = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,4], error= function(e)NA)
})
t = apply(an,1,function(i){
tryCatch(summary(lm(exp[as.character(i[2]),] ~ m[as.character(i[1]),]))$coefficient[2,3], error= function(e)NA)
})
an1 =cbind(an,p)
an1 = cbind(an1,t)
an1$q = p.adjust(as.numeric(an1$p))
summary(lm(exp["MAOB",] ~ m["cg00121904",]$coefficient[2,c(3:4)]
###############################################
______________________________________________
R-help at r-project.org<mailto:R-help at r-project.org> mailing list -- To UNSUBSCRIBE and more, see https://stat.ethz.ch/mailman/listinfo/r-help
PLEASE do read the posting guide http://www.R-project.org/posting-guide.html
and provide commented, minimal, self-contained, reproducible code.