topslam

Will generate a trajectory using topslam.

This method was wrapped inside a container. The original code of this method is available here.

ti_topslam(n_components = 2L, n_neighbors = 10L, linear_dims = 0L,
  max_iters = 1000L, dimreds = c("t-SNE", "PCA", "Spectral", "Isomap",
  "ICA"))

Arguments

n_components

The number of components. Domain: U(2, 10). Default: 2. Format: integer.

n_neighbors

The number of neighbors. Domain: U(2, 100). Default: 10. Format: integer.

linear_dims

. Domain: U(0, 5). Default: 0. Format: integer.

max_iters

The number of iterations to optimize over. Domain: e^U(2.30, 9.21). Default: 1000. Format: integer.

dimreds

Which dimensionality reductions to use and/or ICA. Domain: all subsets of t-SNE, PCA, Spectral, Isomap, ICA. Default: t-SNE, PCA, Spectral, Isomap, ICA. Format: subset.

Value

A TI method wrapper to be used together with infer_trajectory

References

Zwiessele, M., Lawrence, N.D., 2016. Topslam: Waddington Landscape Recovery for Single Cell Experiments.