uSORT: uSORT: A self-refining ordering pipeline for gene selection

Description Usage Arguments Details Value See Also Examples

View source: R/uSORT_main.R

Description

This package is designed to uncover the intrinsic cell progression path from single-cell RNA-seq data.

The main function of uSORT-pacakge which provides a workflow of sorting scRNA-seq data.

Usage

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
uSORT(exprs_file, log_transform = TRUE, remove_outliers = TRUE,
  preliminary_sorting_method = c("autoSPIN", "sWanderlust", "monocle",
  "Wanderlust", "SPIN", "none"), refine_sorting_method = c("autoSPIN",
  "sWanderlust", "monocle", "Wanderlust", "SPIN", "none"),
  project_name = "uSORT", result_directory = getwd(), nCores = 1,
  save_results = TRUE, reproduce_seed = 1234,
  scattering_cutoff_prob = 0.75, driving_force_cutoff = NULL,
  qval_cutoff_featureSelection = 0.05, pre_data_type = c("linear",
  "cyclical"), pre_SPIN_option = c("STS", "neighborhood"),
  pre_SPIN_sigma_width = 1, pre_autoSPIN_alpha = 0.2,
  pre_autoSPIN_randomization = 20, pre_wanderlust_start_cell = NULL,
  pre_wanderlust_dfmap_components = 4, pre_wanderlust_l = 15,
  pre_wanderlust_num_waypoints = 150, pre_wanderlust_waypoints_seed = 2711,
  pre_wanderlust_flock_waypoints = 2, ref_data_type = c("linear",
  "cyclical"), ref_SPIN_option = c("STS", "neighborhood"),
  ref_SPIN_sigma_width = 1, ref_autoSPIN_alpha = 0.2,
  ref_autoSPIN_randomization = 20, ref_wanderlust_start_cell = NULL,
  ref_wanderlust_dfmap_components = 4, ref_wanderlust_l = 15,
  ref_wanderlust_num_waypoints = 150, ref_wanderlust_flock_waypoints = 2,
  ref_wanderlust_waypoints_seed = 2711)

Arguments

exprs_file

Input file name in txt format, with rownames of cells and colnames of genes.

log_transform

Boolean, if log transform the data.

remove_outliers

Boolean, if remove the outliers.

preliminary_sorting_method

Method name for preliminary sorting, including autoSPIN, sWanderlust, monocle, Wanderlust, SPIN, or none.

refine_sorting_method

Method name for refined sorting, including autoSPIN, sWanderlust, monocle, Wanderlust, SPIN, or none.

project_name

A character name as the prefix of the saved result file.

result_directory

The directory indicating where to save the results.

nCores

Number of cores that will be employed for drive gene selection (parallel computing), default is 1.

save_results

Boolean determining if save the results.

reproduce_seed

A seed used for reproducing the result.

scattering_cutoff_prob

Scattering cutoff value probability for gene selection, default 0.75.

driving_force_cutoff

Driving force cutoff value for gene selection, default NULL(automatically).

qval_cutoff_featureSelection

Q value cutoff for gene selection, default 0.05.

pre_data_type

The data type which guides the autoSPIN sorting, including linear, cyclical.

pre_SPIN_option

SPIN contains two options including STS(default), neighborhood.

pre_SPIN_sigma_width

Sigma width parameter for SPIN, default is 1.

pre_autoSPIN_alpha

alpha parameter for autoSPIN, default is 0.2.

pre_autoSPIN_randomization

Number of randomizations for autoSPIN, default is 20.

pre_wanderlust_start_cell

The name of starting cell for wanderlust, default is the first cell from the data.

pre_wanderlust_dfmap_components

Number of components from diffusion map used for wanderlust analysis, default is 4.

pre_wanderlust_l

Number of nearest neighbors used for wanderlust, default is 15.

pre_wanderlust_num_waypoints

Number of waypoint used for wanderlust, default is 150.

pre_wanderlust_waypoints_seed

The seed for reproducing the wanderlust results.

pre_wanderlust_flock_waypoints

The number of times for flocking the waypoints, default is 2.

ref_data_type

The data type which guides the autoSPIN sorting, including linear, cyclical.

ref_SPIN_option

SPIN contains two options including STS(default), neighborhood.

ref_SPIN_sigma_width

Sigma width parameter for SPIN, default is 1.

ref_autoSPIN_alpha

alpha parameter for autoSPIN, default is 0.2.

ref_autoSPIN_randomization

Number of randomizations for autoSPIN, default is 20.

ref_wanderlust_start_cell

The name of starting cell for wanderlust, default is the first cell from the data.

ref_wanderlust_dfmap_components

Number of components from diffusion map used for wanderlust analysis, default is 4.

ref_wanderlust_l

Number of nearest neighbors used for wanderlust, default is 15.

ref_wanderlust_num_waypoints

Number of waypoint used for wanderlust, default is 150

ref_wanderlust_flock_waypoints

The number of times for flocking the waypoints, default is 2.

ref_wanderlust_waypoints_seed

The seed for reproducing the wanderlust results.

Details

This package incorporates data pre-processing, preliminary PCA gene selection, preliminary cell ordering, feature selection, refined cell ordering, and post-analysis interpretation and visualization. The uSORT workflow can be implemented through calling the main function or the GUI. uSORT.

Value

results object (a list)

See Also

uSORT-package, uSORT_GUI

Examples

1
2
3
4
5
6
7
dir <- system.file('extdata', package='uSORT')
file <- list.files(dir, pattern='.txt$', full=TRUE)
#remove the # symbol of the following codes to test
#uSORT_results <- uSORT(exprs_file = file, project_name = "test",
# preliminary_sorting_method = "autoSPIN",
# refine_sorting_method = "sWanderlust",
# save_results = FALSE)

uSORT documentation built on May 31, 2017, 11:39 a.m.