git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@45608 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -56,8 +56,8 @@ snprma2(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verb |
56 | 56 |
the use of clusters or multiple cores (via snow package) to speed up preprocessing. |
57 | 57 |
} |
58 | 58 |
\examples{ |
59 |
-if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
|
60 |
- path <- system.file("celFiles", package="hapmapsnp5") |
|
59 |
+if (require(genomewidesnp6Crlmm) & require(hapmapsnp6) & require(oligoClasses)){ |
|
60 |
+ path <- system.file("celFiles", package="hapmapsnp6") |
|
61 | 61 |
|
62 | 62 |
## the filenames with full path... |
63 | 63 |
## very useful when genotyping samples not in the working directory |
... | ... |
@@ -76,7 +76,7 @@ ocProbesets(50000) |
76 | 76 |
## setup cluster - 8 cores on the machine |
77 | 77 |
setCluster(8, "SOCK") |
78 | 78 |
|
79 |
-path <- system.file("celFiles", package="hapmapsnp5") |
|
79 |
+path <- system.file("celFiles", package="hapmapsnp6") |
|
80 | 80 |
cels <- list.celfiles(path, full.names=TRUE) |
81 | 81 |
snprmaOutput <- snprma2(cels) |
82 | 82 |
|
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@45498 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -1,6 +1,7 @@ |
1 | 1 |
\name{snprma} |
2 | 2 |
\Rdversion{1.1} |
3 | 3 |
\alias{snprma} |
4 |
+\alias{snprma2} |
|
4 | 5 |
|
5 | 6 |
\title{ |
6 | 7 |
Preprocessing tool for SNP arrays. |
... | ... |
@@ -12,6 +13,7 @@ |
12 | 13 |
} |
13 | 14 |
\usage{ |
14 | 15 |
snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns) |
16 |
+snprma2(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns) |
|
15 | 17 |
} |
16 | 18 |
\arguments{ |
17 | 19 |
\item{filenames}{ |
... | ... |
@@ -49,6 +51,10 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
49 | 51 |
\item{mixtureParams}{Parameters from mixture model} |
50 | 52 |
\item{cdfName}{Name of the CDF} |
51 | 53 |
} |
54 |
+\details{ |
|
55 |
+ 'snprma2' allows one to genotype very large datasets (via ff package) and also permits |
|
56 |
+ the use of clusters or multiple cores (via snow package) to speed up preprocessing. |
|
57 |
+ } |
|
52 | 58 |
\examples{ |
53 | 59 |
if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
54 | 60 |
path <- system.file("celFiles", package="hapmapsnp5") |
... | ... |
@@ -60,6 +66,22 @@ if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
60 | 66 |
snprmaOutput[["A"]][1:10,] |
61 | 67 |
snprmaOutput[["B"]][1:10,] |
62 | 68 |
} |
69 |
+\dontrun{ |
|
70 |
+## HPC Example |
|
71 |
+library(ff) |
|
72 |
+library(snow) |
|
73 |
+library(crlmm) |
|
74 |
+## genotype 50K SNPs at a time |
|
75 |
+ocProbesets(50000) |
|
76 |
+## setup cluster - 8 cores on the machine |
|
77 |
+setCluster(8, "SOCK") |
|
78 |
+ |
|
79 |
+path <- system.file("celFiles", package="hapmapsnp5") |
|
80 |
+cels <- list.celfiles(path, full.names=TRUE) |
|
81 |
+snprmaOutput <- snprma2(cels) |
|
82 |
+ |
|
83 |
+} |
|
84 |
+ |
|
63 | 85 |
} |
64 | 86 |
\keyword{manip} |
65 | 87 |
\keyword{classif} |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@45126 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -6,15 +6,12 @@ |
6 | 6 |
Preprocessing tool for SNP arrays. |
7 | 7 |
} |
8 | 8 |
\description{ |
9 |
- |
|
10 | 9 |
SNPRMA will preprocess SNP chips. The preprocessing consists of |
11 | 10 |
quantile normalization to a known target distribution and |
12 |
- summarization to the SNP-Allele level. This function is not |
|
13 |
- typically called directly. |
|
14 |
- |
|
11 |
+ summarization to the SNP-Allele level. |
|
15 | 12 |
} |
16 | 13 |
\usage{ |
17 |
-snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, AFile="A.rda", BFile="B.rda") |
|
14 |
+snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns) |
|
18 | 15 |
} |
19 | 16 |
\arguments{ |
20 | 17 |
\item{filenames}{ |
... | ... |
@@ -36,18 +33,11 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
36 | 33 |
Seed to be used when sampling. |
37 | 34 |
} |
38 | 35 |
\item{cdfName}{ |
39 |
- cdfName: 'genomewidesnp5', 'genomewidesnp6' |
|
40 |
- See \code{validCdfNames}. |
|
36 |
+ cdfName: 'GenomeWideSnp\_5', 'GenomeWideSnp\_6' |
|
37 |
+} |
|
38 |
+ \item{sns}{ |
|
39 |
+ Sample names. |
|
41 | 40 |
} |
42 |
- |
|
43 |
-\item{AFile}{'character' for naming the quantile normalized |
|
44 |
-intensities for the A allele. This object also contains the |
|
45 |
-normalized intensities for the nonpolymorphic loci, depending on the |
|
46 |
-platform } |
|
47 |
- |
|
48 |
-\item{BFile}{'character' for naming the quantile normalized |
|
49 |
-intensities for the B allele. } |
|
50 |
- |
|
51 | 41 |
} |
52 | 42 |
\value{ |
53 | 43 |
\item{A}{Summarized intensities for Allele A} |
... | ... |
@@ -66,11 +56,9 @@ if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
66 | 56 |
## the filenames with full path... |
67 | 57 |
## very useful when genotyping samples not in the working directory |
68 | 58 |
cels <- list.celfiles(path, full.names=TRUE) |
69 |
- snprmaResult <- snprma(cels, cdfName="genomewidesnp5") |
|
70 |
- str(snprmaResult) |
|
71 |
- load("A.rda") |
|
72 |
- A[1:10, ] |
|
73 |
- unlink(list("A.rda", "B.rda")) |
|
59 |
+ snprmaOutput <- snprma(cels) |
|
60 |
+ snprmaOutput[["A"]][1:10,] |
|
61 |
+ snprmaOutput[["B"]][1:10,] |
|
74 | 62 |
} |
75 | 63 |
} |
76 | 64 |
\keyword{manip} |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@45083 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -14,7 +14,7 @@ |
14 | 14 |
|
15 | 15 |
} |
16 | 16 |
\usage{ |
17 |
-snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns, AFile="A.rda", BFile="B.rda") |
|
17 |
+snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, AFile="A.rda", BFile="B.rda") |
|
18 | 18 |
} |
19 | 19 |
\arguments{ |
20 | 20 |
\item{filenames}{ |
... | ... |
@@ -40,10 +40,6 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
40 | 40 |
See \code{validCdfNames}. |
41 | 41 |
} |
42 | 42 |
|
43 |
- \item{sns}{ |
|
44 |
- Sample names. |
|
45 |
-} |
|
46 |
- |
|
47 | 43 |
\item{AFile}{'character' for naming the quantile normalized |
48 | 44 |
intensities for the A allele. This object also contains the |
49 | 45 |
normalized intensities for the nonpolymorphic loci, depending on the |
... | ... |
@@ -70,9 +66,11 @@ if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
70 | 66 |
## the filenames with full path... |
71 | 67 |
## very useful when genotyping samples not in the working directory |
72 | 68 |
cels <- list.celfiles(path, full.names=TRUE) |
73 |
- snprmaOutput <- snprma(cels) |
|
74 |
- snprmaOutput[["A"]][1:10,] |
|
75 |
- snprmaOutput[["B"]][1:10,] |
|
69 |
+ snprmaResult <- snprma(cels, cdfName="genomewidesnp5") |
|
70 |
+ str(snprmaResult) |
|
71 |
+ load("A.rda") |
|
72 |
+ A[1:10, ] |
|
73 |
+ unlink(list("A.rda", "B.rda")) |
|
76 | 74 |
} |
77 | 75 |
} |
78 | 76 |
\keyword{manip} |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@44778 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -6,12 +6,15 @@ |
6 | 6 |
Preprocessing tool for SNP arrays. |
7 | 7 |
} |
8 | 8 |
\description{ |
9 |
+ |
|
9 | 10 |
SNPRMA will preprocess SNP chips. The preprocessing consists of |
10 | 11 |
quantile normalization to a known target distribution and |
11 |
- summarization to the SNP-Allele level. |
|
12 |
+ summarization to the SNP-Allele level. This function is not |
|
13 |
+ typically called directly. |
|
14 |
+ |
|
12 | 15 |
} |
13 | 16 |
\usage{ |
14 |
-snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns) |
|
17 |
+snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns, AFile="A.rda", BFile="B.rda") |
|
15 | 18 |
} |
16 | 19 |
\arguments{ |
17 | 20 |
\item{filenames}{ |
... | ... |
@@ -33,11 +36,22 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
33 | 36 |
Seed to be used when sampling. |
34 | 37 |
} |
35 | 38 |
\item{cdfName}{ |
36 |
- cdfName: 'GenomeWideSnp\_5', 'GenomeWideSnp\_6' |
|
39 |
+ cdfName: 'genomewidesnp5', 'genomewidesnp6' |
|
40 |
+ See \code{validCdfNames}. |
|
37 | 41 |
} |
42 |
+ |
|
38 | 43 |
\item{sns}{ |
39 | 44 |
Sample names. |
40 | 45 |
} |
46 |
+ |
|
47 |
+\item{AFile}{'character' for naming the quantile normalized |
|
48 |
+intensities for the A allele. This object also contains the |
|
49 |
+normalized intensities for the nonpolymorphic loci, depending on the |
|
50 |
+platform } |
|
51 |
+ |
|
52 |
+\item{BFile}{'character' for naming the quantile normalized |
|
53 |
+intensities for the B allele. } |
|
54 |
+ |
|
41 | 55 |
} |
42 | 56 |
\value{ |
43 | 57 |
\item{A}{Summarized intensities for Allele A} |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@43728 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -50,7 +50,7 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
50 | 50 |
\item{cdfName}{Name of the CDF} |
51 | 51 |
} |
52 | 52 |
\examples{ |
53 |
-if (require(genomewidesnp5Crlmm) & require(hapmapsnp5)){ |
|
53 |
+if (require(genomewidesnp5Crlmm) & require(hapmapsnp5) & require(oligoClasses)){ |
|
54 | 54 |
path <- system.file("celFiles", package="hapmapsnp5") |
55 | 55 |
|
56 | 56 |
## the filenames with full path... |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@40309 bc3139a8-67e5-0310-9ffc-ced21a209358
... | ... |
@@ -33,7 +33,7 @@ snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbo |
33 | 33 |
Seed to be used when sampling. |
34 | 34 |
} |
35 | 35 |
\item{cdfName}{ |
36 |
- cdfName: 'GenomeWideSnp_5', 'GenomeWideSnp_6' |
|
36 |
+ cdfName: 'GenomeWideSnp\_5', 'GenomeWideSnp\_6' |
|
37 | 37 |
} |
38 | 38 |
\item{sns}{ |
39 | 39 |
Sample names. |
git-svn-id: file:///home/git/hedgehog.fhcrc.org/bioconductor/trunk/madman/Rpacks/crlmm@38548 bc3139a8-67e5-0310-9ffc-ced21a209358
1 | 1 |
new file mode 100644 |
... | ... |
@@ -0,0 +1,65 @@ |
1 |
+\name{snprma} |
|
2 |
+\Rdversion{1.1} |
|
3 |
+\alias{snprma} |
|
4 |
+ |
|
5 |
+\title{ |
|
6 |
+ Preprocessing tool for SNP arrays. |
|
7 |
+} |
|
8 |
+\description{ |
|
9 |
+ SNPRMA will preprocess SNP chips. The preprocessing consists of |
|
10 |
+ quantile normalization to a known target distribution and |
|
11 |
+ summarization to the SNP-Allele level. |
|
12 |
+} |
|
13 |
+\usage{ |
|
14 |
+snprma(filenames, mixtureSampleSize = 10^5, fitMixture = FALSE, eps = 0.1, verbose = TRUE, seed = 1, cdfName, sns) |
|
15 |
+} |
|
16 |
+\arguments{ |
|
17 |
+ \item{filenames}{ |
|
18 |
+ 'character' vector with file names. |
|
19 |
+} |
|
20 |
+ \item{mixtureSampleSize}{ |
|
21 |
+ Sample size to be use when fitting the mixture model. |
|
22 |
+} |
|
23 |
+ \item{fitMixture}{ |
|
24 |
+ 'logical'. Fit the mixture model? |
|
25 |
+} |
|
26 |
+ \item{eps}{ |
|
27 |
+ Stop criteria. |
|
28 |
+} |
|
29 |
+ \item{verbose}{ |
|
30 |
+ 'logical'. |
|
31 |
+} |
|
32 |
+ \item{seed}{ |
|
33 |
+ Seed to be used when sampling. |
|
34 |
+} |
|
35 |
+ \item{cdfName}{ |
|
36 |
+ cdfName: 'GenomeWideSnp_5', 'GenomeWideSnp_6' |
|
37 |
+} |
|
38 |
+ \item{sns}{ |
|
39 |
+ Sample names. |
|
40 |
+} |
|
41 |
+} |
|
42 |
+\value{ |
|
43 |
+ \item{A}{Summarized intensities for Allele A} |
|
44 |
+ \item{B}{Summarized intensities for Allele B} |
|
45 |
+ \item{sns}{Sample names} |
|
46 |
+ \item{gns}{SNP names} |
|
47 |
+ \item{SNR}{Signal-to-noise ratio} |
|
48 |
+ \item{SKW}{Skewness} |
|
49 |
+ \item{mixtureParams}{Parameters from mixture model} |
|
50 |
+ \item{cdfName}{Name of the CDF} |
|
51 |
+} |
|
52 |
+\examples{ |
|
53 |
+if (require(genomewidesnp5Crlmm) & require(hapmapsnp5)){ |
|
54 |
+ path <- system.file("celFiles", package="hapmapsnp5") |
|
55 |
+ |
|
56 |
+ ## the filenames with full path... |
|
57 |
+ ## very useful when genotyping samples not in the working directory |
|
58 |
+ cels <- list.celfiles(path, full.names=TRUE) |
|
59 |
+ snprmaOutput <- snprma(cels) |
|
60 |
+ snprmaOutput[["A"]][1:10,] |
|
61 |
+ snprmaOutput[["B"]][1:10,] |
|
62 |
+} |
|
63 |
+} |
|
64 |
+\keyword{manip} |
|
65 |
+\keyword{classif} |