Biology Reference
In-Depth Information
265768_at, 1st time series
9.0
observed
estimated
prediction
8.5
filtering
8.0
7.5
7.0
smoothing
6.5
1
2
3
4
5
6
7
8
9
10
11
265768_at, 2nd time series
9.0
observed
estimated
prediction
8.5
filtering
8.0
7.5
7.0
smoothing
6.5
1
2
3
4
5
6
7
8
9
10
11
Fig. 4.2
Observed and estimated expression levels for two time series available for gene
265768 at
in the LASSO model from Sect.
3.5.2
> x = arth12[1:(nrow(arth12) - 2), ]
> y = arth12[-(1:2), "265768_at"]
> lasso.fit = lars(y = y, x = x, type = "lasso")
If we tune the model to find the optimal value for the
L
1
penalty, we can then es-
timate the expression levels of the gene
265768 at
for all past and present time
points.
> lasso.cv = cv.lars(y = y, x = x, mode = "fraction")
> frac = lasso.cv$index[which.min(lasso.cv$cv)]
Search WWH ::
Custom Search