Binary matrix with a row for each clone and column for
each sample. "1" indicates outlier, 0 otherwise.
pred.obs.out
Matrix with a row for each clone and column for
each sample. The entries are the median value for the state with
outliers exceluded for all clones but outliers. The value if the
observed value for the outliers.
pred.out
Matrix with a row for each clone and column for each
sample. The entries are the median value for the state
Details
The outliers are the clones that are dissimilar enough from the clones
assigned to the same state. Magnitude of the factor determines how
many MADs away from a median a value needs to be to be declared an
outlier. Outliers consitent over many samples may indicate
technological artificat with that clone or possibly copy number
polymorpism.
References
Application of Hidden Markov Models to the analysis of the
array CGH data, Fridlyand et.al., JMVA, 2004