Multi-platform scan using projected chi-square

Share:

Description

Scans multi-platform data for all variable length windows have projected chi-square statistc higher than a fixed user-set threshold.

Usage

1
mpscan(y, pos, anchors, win)

Arguments

y

A vector of intensity levels for each platform. y[[k]] should be the intensity levels corresponding to pos[[k]] for platform k.

pos

A vector of sorted integer arrays, one array for each platform. pos[[k]] should give the positions, in increasing order, of the probes of the k-th platform.

anchor

The anchor set, returned by a call to merge.pos(...)

win

Largest allowed window size.

MIN.SNPs

Minimum number of SNPs allowed in a region.

WCHISQ.CUTOFF

Threshold level for the weighted chi-square statistic.

Value

yhat

A vector of fitted y values, yhat[[k]] contains the fitted values to y[[k]].

chpts

A list of change-points, there are shared across platforms.

platform.chisq

The contribution of each platform to the overall chi-square value for each interval.

Z

The Z matrix.

Author(s)

Nancy R. Zhang

References

Zhang, NR, Senbabaoglu, Y. and Li, J.Z. (2009) Joint Estimation of DNA Copy Number from Multiple Platforms. Under review, download manuscript from http://www-stat.stanford.edu/~nzhang/web_multiplatform/

See Also

mpcbs, plot.crossplatform, merge.pos

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
data(mpcbs.example)

# there are 3 platforms represented in this data example: Illumina, Affymetrix, and Agilent.
names(mpcbs.example)

# K is the number of platforms.
K=3

# Store the chromosome positions in vector pos,
# the intensities in vector y:

pos=vector("list",K)
pos[[1]] = mpcbs.example$illu[,1]
pos[[2]] = mpcbs.example$affy[,1]
pos[[3]] = mpcbs.example$agil[,1]

y = vector("list",K)
y[[1]] = mpcbs.example$illu[,2]
y[[2]] = mpcbs.example$affy[,2]
y[[3]] = mpcbs.example$agil[,2]

# Names of the platforms:
platform.names=c("Illumina","Affymetrix","Agilent")

# Get the anchor set.
anchor = merge.pos(pos)