R help - Jun 2012 - Analyzing large files faster

I'm trying to analyze the following data set (sample):

"ID"	                        "adj.P.Val"	"logFC"	 
"Gene.symbol"
"1419156_at"	"5.32e-12"	"2.6462565"
"Sox4"
"1433575_at"	"5.32e-12"	"3.9417089"
"Sox4"
"1428942_at"	"2.64e-11"	"3.9163618"
"Mt2"
"1454699_at"	"2.69e-10"	"1.8654677"
"LOC100047324///Sesn1"
"1416926_at"	"3.19e-10"	"2.172342"
"Trp53inp1"
"1422557_s_at"	"1.58e-09"	"2.9569254"
"Mt1"
etc.

using the following code:

muscle = read.table(file="/Users/bob/Desktop/Muscle/musclesmall.txt",
header
= TRUE, colClasses = "character", fill = TRUE)

upregulated_list = c()
downregulated_list = c()
nochange = c()
p_thresh = 6.51e-06
x=1

while (x <= nrow(muscle)) {
this_pval = muscle[x,"adj.P.Val"]
this_M = muscle[x, "logFC"]

if (muscle[x, "Gene.symbol"] == "") {
	x= x +1
	} 
	else {if ((this_M >= 1.0) & (this_pval <= p_thresh)) {
	upregulated_list <- append(upregulated_list,
muscle[x,"Gene.symbol"],after=length(upregulated_list))
	x = x +1}
	else {if ((this_M <= -1) & (this_pval <= p_thresh)) {
		downregulated_list <- append(downregulated_list,
muscle[x,"Gene.symbol"],after=length(downregulated_list))
		x = x+1
		}
		  
		 else {if ((this_M > -1) & (this_M < 1)) {
		 		nochange <- append(nochange,
muscle[x,"Gene.symbol"],after=length(nochange))
		 		x = x+1}
		 		}
		 		}
		 		}
		 	}

This process, however, goes line-by-line and the data has 22,000 rows, so
running the process takes an enormous amount of time. Is there any way for
me to do the analysis faster?

--
View this message in context:
http://r.789695.n4.nabble.com/Analyzing-large-files-faster-tp4633165.html
Sent from the R help mailing list archive at Nabble.com.

Hello,

The trick is to use index vectors. They allow us to do without loops.

Try the following.


muscle <- read.table(text='
"ID"	            "adj.P.Val"	"logFC"	      
"Gene.symbol"
"1419156_at"	"5.32e-12"	"2.6462565"
"Sox4"
"1433575_at"	"5.32e-12"	"3.9417089"
"Sox4"
"1428942_at"	"2.64e-11"	"3.9163618"
"Mt2"
"1454699_at"	"2.69e-10"	"1.8654677"
"LOC100047324///Sesn1"
"1416926_at"	"3.19e-10"	"2.172342"
"Trp53inp1"
"1422557_s_at"	"1.58e-09"	"2.9569254"
"Mt1"
', header=TRUE, stringsAsFactors=FALSE)

muscle

p_thresh = 6.51e-06

# Create index vectors
gsym <- muscle$Gene.symbol != ""
this_pval <- muscle$adj.P.Val <= p_thresh
this_Ma <- muscle$logFC > -1
this_Mb <- muscle$logFC < 1

# Use them
downregulated_list <- muscle$Gene.symbol[gsym & !this_Ma & this_pval]
upregulated_list <- muscle$Gene.symbol[gsym & !this_Mb & this_pval]
nochange <- muscle$Gene.symbol[gsym & this_Ma & this_Mb]

# See the result [ Maybe with head() ]
upregulated_list
downregulated_list
nochange


Hope this helps,

Rui Barradas
Em 12-06-2012 21:55, mousy0815 escreveu:
> upregulated_list = c()
> downregulated_list = c()
> nochange = c()
> p_thresh = 6.51e-06
> x=1
>
> while (x <= nrow(muscle)) {
> this_pval = muscle[x,"adj.P.Val"]
> this_M = muscle[x, "logFC"]