autoCluster.batch: Cluster the preprocessed fcs files from different studies in...

Description Usage Arguments Value Examples

View source: R/autoCluster.batch.R

Description

A function that clusters the pre-processed fcs files from different studies in batch.

Usage

1
2
3
autoCluster.batch(preprocessOutputFolder,
  excludeClusterParameters = c("TIME"), labelQuantile = 0.95,
  clusterFunction = flowSOM.MC, minPercent = 0.05, ...)

Arguments

preprocessOutputFolder

Directory where the preprocessed results are stored. Should be the same with the outpath argument in preprocessing.batch function.

excludeClusterParameters

A vector specifying the name of markers not to be used for clustering and labeling. Typical example includes: Time, cell_length.

labelQuantile

A number between 0.5 and 1. Used to specify the minimum percent of cells in a cluster required to express higher or lower level of a marker than the cutoff value for labeling.

clusterFunction

The name of unsupervised clustering function the user wish to use for clustering the cells. The default is "flowSOM.MC". The first argument of the function must take a flow frame, the second argument of the function must take a vector of excludeClusterParameters. The function must return a list of clusters containing cell IDs. flowSOM.MC and flowHC are implemented in the package. For other methods, please make your own wrapper functions.

minPercent

A number between 0 and 0.5. Used to specify the minimum percent of cells in the positive and negative region after bisection. Keep it small to avoid bisecting uni-mode distributions.

...

Pass arguments to clusterFunction

Value

A vector of labels identified in the cytometry data.

Examples

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
#get meta-data
fn=system.file("extdata","fcs_info.csv",package="MetaCyto")
fcs_info=read.csv(fn,stringsAsFactors=FALSE,check.names=FALSE)
fcs_info$fcs_files=system.file("extdata",fcs_info$fcs_files,
                               package="MetaCyto")
# Make sure the transformation parameter "b" and the "assay" argument
# are correct of FCM and CyTOF files
b=assay=rep(NA,nrow(fcs_info))
b[grepl("CyTOF",fcs_info$study_id)]=1/8
b[grepl("FCM",fcs_info$study_id)]=1/150
assay[grepl("CyTOF",fcs_info$study_id)]="CyTOF"
assay[grepl("FCM",fcs_info$study_id)]="FCM"
# preprocessing
preprocessing.batch(inputMeta=fcs_info,
                    assay=assay,
                    b=b,
                    outpath="Example_Result/preprocess_output",
                    excludeTransformParameters=c("FSC-A","FSC-W","FSC-H",
                    "Time","Cell_length"))
# Make sure marker names are consistant in different studies
files=list.files("Example_Result",pattern="processed_sample",
                 recursive=TRUE,full.names=TRUE)
nameUpdator("CD8B","CD8",files)
# find the clusters
excludeClusterParameters=c("FSC-A","FSC-W","FSC-H","SSC-A",
                           "SSC-W","SSC-H","Time",
                          "CELL_LENGTH","DEAD","DNA1","DNA2")
cluster_label=autoCluster.batch(
              preprocessOutputFolder="Example_Result/preprocess_output",
              excludeClusterParameters=excludeClusterParameters,
              labelQuantile=0.95,
              clusterFunction=flowHC)

MetaCyto documentation built on Nov. 8, 2020, 7:50 p.m.