Calculating feature importances across trajectories

Uses the feature importance measures of ranger or caret. calculate_overall_feature_importance calculates the importance for the whole trajectory, calculate_milestone_feature_importance calculates it for individual milestones (eg. branching points)

calculate_milestone_feature_importance(traj, expression_source = "expression",
  milestones_oi = NULL, method = "ranger", method_params = list())

calculate_overall_feature_importance(traj, expression_source = "expression",
  method = "ranger", method_params = list())

calculate_waypoint_feature_importance(traj, expression_source = "expression",
  waypoints = NULL, method = "ranger", method_params = list())

calculate_cell_feature_importance(traj, expression_source = "expression",
  method = "ranger", method_params = list())

calculate_branch_feature_importance(traj, expression_source = "expression",
  method = "ranger", method_params = list())

calculate_branching_point_feature_importance(traj,
  expression_source = "expression", milestones_oi = traj$milestone_ids,
  method = "ranger", method_params = list())

Arguments

traj	A trajectory object containing expression values and a trajectory.
expression_source	The expression_source, if not provided will use the expression within the trajectory
milestones_oi	The milestone(s) for which to calculate feature importance
method	The method to do regressions, can be `ranger` or any regression model from caret
method_params	Parameters given to the method
waypoints	The waypoints, optional