Predict the principal components for new trajectories

Predict the principal components for new trajectories

# S3 method for fmca
predict(
  object,
  newdata = NULL,
  method = c("precompute", "parallel"),
  verbose = TRUE,
  nCores = max(1, ceiling(detectCores()/2)),
  ...
)

Arguments

object

output of compute_optimal_encoding function.

newdata

data.frame containing id, id of the trajectory, time, time at which a change occurs and state, associated state. All individuals must begin at the same time T0 and end at the same time Tmax (use cut_data).

method

computation method: "parallel" or "precompute": precompute all integrals (efficient when the number of unique time values is low)

verbose

if TRUE print some information

nCores

number of cores used for parallelization (only if method == "parallel"). Default is half the cores.

...

parameters for integrate function (see details).

Value

principal components for the individuals

See also

Other encoding functions: compute_optimal_encoding(), get_encoding(), plot.fmca(), plotComponent(), plotEigenvalues(), print.fmca(), summary.fmca()

Author

Quentin Grimonprez

Examples

# Simulate the Jukes-Cantor model of nucleotide replacement
K <- 4
Tmax <- 6
PJK <- matrix(1 / 3, nrow = K, ncol = K) - diag(rep(1 / 3, K))
lambda_PJK <- c(1, 1, 1, 1)
d_JK <- generate_Markov(
  n = 10, K = K, P = PJK, lambda = lambda_PJK, Tmax = Tmax,
  labels = c("A", "C", "G", "T")
)
d_JK2 <- cut_data(d_JK, Tmax)

# create basis object
m <- 6
b <- create.bspline.basis(c(0, Tmax), nbasis = m, norder = 4)
# \donttest{
# compute encoding
encoding <- compute_optimal_encoding(d_JK2, b, computeCI = FALSE, nCores = 1)
#> ######### Compute encoding #########
#> Number of individuals: 10
#> Number of states: 4
#> Basis type: bspline
#> Number of basis functions: 6
#> Number of cores: 1
#> 
  |                                                  | 0 % elapsed=00s   
  |=========                                         | 17% elapsed=00s, remaining~00s
  |=================                                 | 33% elapsed=00s, remaining~00s
  |=========================                         | 50% elapsed=00s, remaining~00s
  |==================================                | 67% elapsed=00s, remaining~00s
  |==========================================        | 83% elapsed=00s, remaining~00s
  |==================================================| 100% elapsed=00s, remaining~00s
#> 
#> DONE in 0.17s
#> ---- Compute U matrix:
#> 
  |                                                  | 0 % elapsed=00s   
  |===                                               | 5 % elapsed=00s, remaining~01s
  |=====                                             | 10% elapsed=00s, remaining~01s
  |========                                          | 14% elapsed=00s, remaining~01s
  |==========                                        | 19% elapsed=00s, remaining~01s
  |============                                      | 24% elapsed=00s, remaining~01s
  |===============                                   | 29% elapsed=00s, remaining~01s
  |=================                                 | 33% elapsed=00s, remaining~01s
  |====================                              | 38% elapsed=00s, remaining~00s
  |======================                            | 43% elapsed=00s, remaining~00s
  |========================                          | 48% elapsed=00s, remaining~00s
  |===========================                       | 52% elapsed=00s, remaining~00s
  |=============================                     | 57% elapsed=00s, remaining~00s
  |===============================                   | 62% elapsed=00s, remaining~00s
  |==================================                | 67% elapsed=01s, remaining~00s
  |====================================              | 71% elapsed=01s, remaining~00s
  |=======================================           | 76% elapsed=01s, remaining~00s
  |=========================================         | 81% elapsed=01s, remaining~00s
  |===========================================       | 86% elapsed=01s, remaining~00s
  |==============================================    | 90% elapsed=01s, remaining~00s
  |================================================  | 95% elapsed=01s, remaining~00s
  |==================================================| 100% elapsed=01s, remaining~00s
#> 
#> DONE in 0.9s
#> ---- Compute encoding: 
#> DONE in 0s
#> Run Time: 1.08s

# predict principal components
d_JK_predict <- generate_Markov(
  n = 5, K = K, P = PJK, lambda = lambda_PJK, Tmax = Tmax,
  labels = c("A", "C", "G", "T")
)
d_JK_predict2 <- cut_data(d_JK, Tmax)

pc <- predict(encoding, d_JK_predict2, nCores = 1)
#> ######### Predict Principal Components #########
#> 
  |                                                  | 0 % elapsed=00s   
  |=========                                         | 17% elapsed=00s, remaining~00s
  |=================                                 | 33% elapsed=00s, remaining~00s
  |=========================                         | 50% elapsed=00s, remaining~00s
  |==================================                | 67% elapsed=00s, remaining~00s
  |==========================================        | 83% elapsed=00s, remaining~00s
  |==================================================| 100% elapsed=00s, remaining~00s
#> 
#> DONE in 0.14s
# }