PHATE (Potential of Heat-diffusion for Affinity-based Transition Embedding) is a tool for visualizing high dimensional data.
Changeset:
cluster
submodule
cluster.kmeans
function to perform KMeans on the diffusion potentialdpi
argument to plot.rotate_scatter3d
Changeset:
Changes:
tasklogger
for logging; fixes issues with Windows 7 verbose outputscprep
in tutorials for simpler analysisphate.plot.scatter
now takes keyword arguments x
, y
and z
rather than a list-likephate.plot.scatter
has keyword arguments for axis label prefix label_prefix
(as an alternative to labelling axes individually), plot title title
, and legend location legend_loc
PHATE v0.2.8 adds plotting utilities:
phate.plot.scatter2d
: 2D scatterplotphate.plot.scatter3d
: 3D scatterplotphate.plot.rotate_scatter3d
: rotating 3D scatterplot (gif or mp4)All plotting functions accept either data or a PHATE object as input. The color vector c
can be continuous or categorical, and need not be numeric. Legends / colorbars are generated with randomized point order on the canvas.
PHATE now implements a sparse, fast alpha decay kernel which has minimal memory requirements, compared to the old alpha decay which required a parwise distance matrix. Alpha decay now runs by default. In order to use the k nearest neighbors kernel, run with a=None
.
Other notable changes:
graphtools
implementation of kernel matricesgamma
replaces potential_method
for the selection of informational distances. gamma=1
is equivalent to a log potential (default) and gamma=0
is equivalent to a square root potentialalpha_decay
and potential_method
are deprecated.The Python version of PHATE now accepts both distance matrices and affinity matrices with the keyword knn_dist='precomputed'
.
We assume distance matrices have only zeroes along the diagonal, and affinity matrices have no zeroes on the diagonal.
PHATE now accepts scanpy's native AnnData format
Version 2.0 implements fast scalable PHATE in Python (2.7, >=3.5), MATLAB and R.
PHATE now runs in seconds to minutes on tens of thousands of cells. Benchmarking shows runtime of ~3 hours on >1,000,000 cells.
Key changes: