Branched expression analysis modeling (BEAM).

Description

Identify genes with branch-dependent expression. Branches in single-cell trajectories are generated by cell fate decisions in development and also arise when analyzing genetic, chemical, or environmental perturbations. Branch expression analysis modeling is a statistical approach for finding genes that are regulated in a manner that depends on the branch. Consider a progenitor cell that generates two distinct cell types. A single-cell trajectory that includes progenitor cells and both differentiated cell types will capture the "decision" as a branch point, with progenitors upstream of the branch and the differentiated cells positioned along distinct branches. These branches will be characterized by distinct gene expression programs. BEAM aims to find all genes that differ between the branches. Such "branch-dependent" genes can help identify the mechanism by which the fate decision is made. BEAM() Takes a CellDataSet and either a specified branch point, or a pair of trajectory outcomes (as States). If a branch point is provided, the function returns a dataframe of test results for dependence on that branch. If a pair of outcomes is provided, it returns test results for the branch that unifies those outcomes into a common path to the trajectory's root state. BEAM() compares two models with a likelihood ratio test for branch-dependent expression. The full model is the product of smooth Pseudotime and the Branch a cell is assigned to. The reduced model just includes Pseudotime. You can modify these to include arbitrary additional effects in the full or both models.

Usage

BEAM(cds, fullModelFormulaStr = "~sm.ns(Pseudotime, df = 3)*Branch",
  reducedModelFormulaStr = "~sm.ns(Pseudotime, df = 3)",
  branch_states = NULL, branch_point = 1, relative_expr = TRUE,
  branch_labels = NULL, verbose = FALSE, cores = 1, ...)

Arguments

Value

a data frame containing the p values and q-values from the BEAM test, with one row per gene.

The CellDataSet class

Description

The main class used by Monocle to hold single cell expression data. CellDataSet extends the basic Bioconductor ExpressionSet class.

Details

This class is initialized from a matrix of expression values Methods that operate on CellDataSet objects constitute the basic Monocle workflow.

Methods for the CellDataSet class

Description

Methods for the CellDataSet class

Usage

list(list("sizeFactors"), list("CellDataSet"))(object)
list(list("sizeFactors"), list("CellDataSet,numeric"))(object) <- value
list(list("estimateSizeFactors"), list("CellDataSet"))(object, locfunc = median, ...)
list(list("estimateDispersions"), list("CellDataSet"))(object, modelFormulaStr = "~ 1",
  relative_expr = TRUE, min_cells_detected = 1, remove_outliers = TRUE,
  cores = 1, ...)

Arguments

The CellType class

Description

Classifies cells using a criterion function.

Details

Classifies cells via a user-defined gating function. The gating function accepts as input the entire matrix of expression data from a CellDataSet , and return TRUE or FALSE for each cell in it, depending on whether each meets the criteria in the gating function

The CellTypeHierarchy class

Description

Classifies cells according to a hierarchy of types.

Details

Classifies cells according to a hierarchy of types via user-defined gating functions.

Add a new cell type

Description

adds a cell type to a pre-existing CellTypeHierarchy and produces a function that accepts expression data from a CellDataSet. When the function is called on a CellDataSet a boolean vector is returned that indicates whether each cell is or is not the cell type that was added by addCellType.

Usage

addCellType(cth, cell_type_name, classify_func,
  parent_cell_type_name = "root")

Arguments

Test for branch-dependent expression

Description

Testing for branch-dependent expression with BEAM first involves constructing a CellDataSet that assigns each cell to a branch, and then performing a likelihood ratio test to see if the branch assignments significantly improves the fit over a null model that does not split the cells. branchTest() implements these two steps.

Usage

branchTest(cds, fullModelFormulaStr = "~sm.ns(Pseudotime, df = 3)*Branch",
  reducedModelFormulaStr = "~sm.ns(Pseudotime, df = 3)",
  branch_states = NULL, branch_point = 1, relative_expr = TRUE,
  cores = 1, branch_labels = NULL, verbose = FALSE, ...)

Arguments

Value

a data frame containing the p values and q-values from the likelihood ratio tests on the parallel arrays of models.

Build a CellDataSet that splits cells among two branches

Description

Analyzing branches with BEAM requires fitting two models to the expression data for each gene. The full model assigns each cell to one of the two outcomes of the branch, and the reduced model excludes this assignment. buildBranchBranchCellDataSet() takes a CellDataSet object and returns a version where the cells are assigned to one of two branches. The branch for each cell is encoded in a new column, "Branch", in the pData table in the returned CellDataSet.

Usage

buildBranchCellDataSet(cds, progenitor_method = c("sequential_split",
  "duplicate"), branch_states = NULL, branch_point = 1,
  branch_labels = NULL, stretch = TRUE)

Arguments

Value

a CellDataSet with the duplicated cells and stretched branches

Compute the area between curves (ABC) for branch-dependent genes

Description

This function is used to calculate the ABC score based on the the nature spline curves fitted for each branch. ABC score is used to quantify the total magnitude of divergence between two branchs. By default, the ABC score is the area between two fitted spline curves. The ABC score can be used to rank gene divergence. When coupled with p-val calculated from the branchTest, it can be used to identify potential major regulators for branch bifurcation.

Usage

calABCs(cds, trend_formula = "~sm.ns(Pseudotime, df = 3)*Branch",
  branch_point = 1, trajectory_states = NULL, relative_expr = TRUE,
  stretch = TRUE, cores = 1, verbose = F, min_expr = 0.5,
  integer_expression = FALSE, num = 5000, branch_labels = NULL, ...)

Arguments

Value

a data frame containing the ABCs (Area under curves) score as the first column and other meta information from fData

Calculate the Instantaneous Log Ratio between two branches

Description

This function is used to calculate the Instant Log Ratio between two branches which can be used to prepare the heatmap demonstrating the branch gene expression divergence hirearchy. If "stretch" is specifified, each branch will be firstly stretched into maturation level from 0-100. Since the results when we use "stretching" are always better and IRLs for non-stretched spline curves are often mismatched, we may only turn down "non-stretch" functionality in future versions. Then, we fit two separate nature spline curves for each individual linages. The log-ratios of the value on each spline curve corresponding to each branch are calculated, which can be used as a measure for the magnitude of divergence between two branching branchs.

Usage

calILRs(cds, trend_formula = "~sm.ns(Pseudotime, df = 3)*Branch",
  branch_point = 1, trajectory_states = NULL, relative_expr = TRUE,
  stretch = TRUE, cores = 1, ILRs_limit = 3, label_by_short_name = TRUE,
  useVST = FALSE, round_exprs = FALSE, output_type = "all",
  branch_labels = NULL, file = NULL, return_all = F, verbose = FALSE,
  ...)

Arguments

Value

a ggplot2 plot object

Calibrate_per_cell_total_proposal

Description

Calibrate_per_cell_total_proposal

Usage

calibrate_per_cell_total_proposal(relative_exprs_matrix, t_estimate,
  expected_capture_rate, method = c("num_genes", "tpm_fraction"))

Arguments

Get the matrix of pairwise distances between cells

Description

Retrieves a matrix capturing distances between each cell used during cell ordering.

Usage

cellPairwiseDistances(cds)

Arguments

Value

A square, symmetric matrix containing the distances between each cell in the reduced-dimensionality space.

Examples

D <- cellPairwiseDistances(HSMM)

Sets the matrix containing distances between each pair of cells used by Monocle during cell ordering. Not intended to be called directly.

Description

Sets the matrix containing distances between each pair of cells used by Monocle during cell ordering. Not intended to be called directly.

Usage

cellPairwiseDistances(cds) <- value

Arguments

Value

An updated CellDataSet object

Examples

cds <- cellPairwiseDistances(D)

Cluster cells into a specified number of groups based on .

Description

Unsupervised clustering of cells is a common step in many single-cell expression workflows. In an experiment containing a mixture of cell types, each cluster might correspond to a different cell type. This method takes a CellDataSet as input along with a requested number of clusters, clusters them with an unsupervised algorithm (by default, density peak clustering), and then returns the CellDataSet with the cluster assignments stored in the pData table. When number of clusters is set to NULL (num_clusters = NULL), the decision plot as introduced in the reference will be plotted and the users are required to check the decision plot to select the rho and delta to determine the number of clusters to cluster. When the dataset is big, for example > 50 k, we recommend the user to use the Louvain clustering algorithm which is inspired from phenograph paper. Note Louvain doesn't support the num_cluster argument but the k (number of k-nearest neighbors) is relevant to the final clustering number. The implementation of Louvain clustering is based on the Rphenograph package but updated based on our requirement (for example, changed the jaccard_coeff function as well as adding louvain_iter argument, etc.)

Usage

clusterCells(cds, skip_rho_sigma = F, num_clusters = NULL,
  inspect_rho_sigma = F, rho_threshold = NULL, delta_threshold = NULL,
  peaks = NULL, gaussian = T, cell_type_hierarchy = NULL,
  frequency_thresh = NULL, enrichment_thresh = NULL,
  clustering_genes = NULL, k = 50, louvain_iter = 1, weight = FALSE,
  method = c("densityPeak", "louvain", "DDRTree"), verbose = F, ...)

Arguments

Value

an updated CellDataSet object, in which phenoData contains values for Cluster for each cell

References

Rodriguez, A., & Laio, A. (2014). Clustering by fast search and find of density peaks. Science, 344(6191), 1492-1496. doi:10.1126/science.1242072

Vincent D. Blondel, Jean-Loup Guillaume, Renaud Lambiotte, Etienne Lefebvre: Fast unfolding of communities in large networks. J. Stat. Mech. (2008) P10008

Jacob H. Levine and et.al. Data-Driven Phenotypic Dissection of AML Reveals Progenitor-like Cells that Correlate with Prognosis. Cell, 2015.

Clusters genes by pseudotime trend.

Description

This function takes a matrix of expression values and performs k-means clustering on the genes.

Usage

clusterGenes(expr_matrix, k, method = function(x) {     as.dist((1 -
  cor(Matrix::t(x)))/2) }, ...)

Arguments

Value

a pam cluster object

Examples

full_model_fits <- fitModel(HSMM[sample(nrow(fData(HSMM_filtered)), 100),],
modelFormulaStr="~sm.ns(Pseudotime)")
expression_curve_matrix <- responseMatrix(full_model_fits)
clusters <- clusterGenes(expression_curve_matrix, k=4)
plot_clusters(HSMM_filtered[ordering_genes,], clusters)

Compare model fits

Description

Performs likelihood ratio tests on nested vector generalized additive models

Usage

compareModels(full_models, reduced_models)

Arguments

Value

a data frame containing the p values and q-values from the likelihood ratio tests on the parallel arrays of models.

Calculate divergence times for branch-dependent genes

Description

Branch-dependent genes may diverge at different points in pseudotime. detectBifurcationPoint() calculates these times. Although the branch times will be shaped by and distributed around the branch point in the trajectory, upstream regulators tend to branch earlier in pseudotime than their targets.

Usage

detectBifurcationPoint(str_log_df = NULL, ILRs_threshold = 0.1,
  detect_all = T, cds = cds, Branch = "Branch", branch_point = NULL,
  branch_states = c(2, 3), stretch = T, cores = 1,
  trend_formula = "~sm.ns(Pseudotime, df = 3)", ILRs_limit = 3,
  relative_expr = TRUE, label_by_short_name = TRUE, useVST = FALSE,
  round_exprs = FALSE, output_type = "all", return_cross_point = T,
  file = "bifurcation_heatmap", verbose = FALSE, ...)

Arguments

Value

a vector containing the time for the bifurcation point with gene names for each value

Detects genes above minimum threshold.

Description

Sets the global expression detection threshold to be used with this CellDataSet. Counts how many cells each feature in a CellDataSet object that are detectably expressed above a minimum threshold. Also counts the number of genes above this threshold are detectable in each cell.

Usage

detectGenes(cds, min_expr = NULL)

Arguments

Value

an updated CellDataSet object

Examples

HSMM <- detectGenes(HSMM, min_expr=0.1)

Helper function for parallel differential expression testing

Description

test

Usage

diff_test_helper(x, fullModelFormulaStr, reducedModelFormulaStr,
  expressionFamily, relative_expr, weights, disp_func = NULL,
  verbose = FALSE)

Arguments

Test genes for differential expression

Description

Tests each gene for differential expression as a function of pseudotime or according to other covariates as specified. differentialGeneTest is Monocle's main differential analysis routine. It accepts a CellDataSet and two model formulae as input, which specify generalized lineage models as implemented by the VGAM package.

Usage

differentialGeneTest(cds, fullModelFormulaStr = "~sm.ns(Pseudotime, df=3)",
  reducedModelFormulaStr = "~1", relative_expr = TRUE, cores = 1,
  verbose = FALSE)

Arguments

Value

a data frame containing the p values and q-values from the likelihood ratio tests on the parallel arrays of models.

Seealso

vglm

Retrieve a table of values specifying the mean-variance relationship

Description

Calling estimateDispersions computes a smooth function describing how variance in each gene's expression across cells varies according to the mean. This function only works for CellDataSet objects containing count-based expression data, either transcripts or reads.

Usage

dispersionTable(cds)

Arguments

Value

A data frame containing the empirical mean expression, empirical dispersion, and the value estimated by the dispersion model.

Helper function to estimate dispersions

Description

Helper function to estimate dispersions

Usage

estimateDispersionsForCellDataSet(cds, modelFormulaStr, relative_expr,
  min_cells_detected, removeOutliers, verbose = FALSE)

Arguments

Function to calculate the size factor for the single-cell RNA-seq data

@importFrom stats median

Description

Function to calculate the size factor for the single-cell RNA-seq data

@importFrom stats median

Usage

estimateSizeFactorsForMatrix(counts, locfunc = median, round_exprs = TRUE,
  method = "mean-geometric-mean-total")

Arguments

Find the most commonly occuring relative expression value in each cell

Description

Converting relative expression values to mRNA copies per cell requires knowing the most commonly occuring relative expression value in each cell This value typically corresponds to an RPC value of 1. This function finds the most commonly occuring (log-transformed) relative expression value for each column in the provided expression matrix.

Usage

estimate_t(relative_expr_matrix, relative_expr_thresh = 0.1)

Arguments

Details

This function estimates the most abundant relative expression value (t^) using a gaussian kernel density function. It can also optionally output the t^ based on a two gaussian mixture model based on the smsn.mixture from mixsmsn package

Value

an vector of most abundant relative_expr value corresponding to the RPC 1.

Examples

HSMM_fpkm_matrix <- exprs(HSMM)
t_estimate = estimate_t(HSMM_fpkm_matrix)

Export a monocle CellDataSet object to other popular single cell analysis toolkit.

Description

This function takes a monocle CellDataSet and converts it to another type of object used in another popular single cell analysis toolkit. It currently supports Scran and Seurat packages.

Usage

exportCDS(monocle_cds, export_to = c("Seurat", "Scater"),
  export_all = FALSE)

Arguments

Value

a new object in the format of another package, as described in the export_to argument.

Examples

lung <- load_lung()
seurat_lung <- exportCDS(lung)
seurat_lung_all <- exportCDS(lung, export_all = T)
scater_lung <- exportCDS(lung, export_to = 'Scater')
scater_lung_all <- exportCDS(lung, export_to = 'Scater', export_all = T)

Extract a linear ordering of cells from a PQ tree

Description

Extract a linear ordering of cells from a PQ tree

Usage

extract_good_branched_ordering(orig_pq_tree, curr_node, dist_matrix,
  num_branches, reverse_main_path = FALSE)

Arguments

Fits a model for each gene in a CellDataSet object.

Description

This function fits a vector generalized additive model (VGAM) from the VGAM package for each gene in a CellDataSet. By default, expression levels are modeled as smooth functions of the Pseudotime value of each cell. That is, expression is a function of progress through the biological process. More complicated formulae can be provided to account for additional covariates (e.g. day collected, genotype of cells, media conditions, etc).

Usage

fitModel(cds, modelFormulaStr = "~sm.ns(Pseudotime, df=3)",
  relative_expr = TRUE, cores = 1)

Arguments

Details

This function fits a vector generalized additive model (VGAM) from the VGAM package for each gene in a CellDataSet. By default, expression levels are modeled as smooth functions of the Pseudotime value of each cell. That is, expression is a function of progress through the biological process. More complicated formulae can be provided to account for additional covariates (e.g. day collected, genotype of cells, media conditions, etc).

Value

a list of VGAM model objects

Helper function for parallel VGAM fitting

Description

test

Usage

fit_model_helper(x, modelFormulaStr, expressionFamily, relative_expr,
  disp_func = NULL, verbose = FALSE, ...)

Arguments

Fit smooth spline curves and return the residuals matrix

Description

This function will fit smooth spline curves for the gene expression dynamics along pseudotime in a gene-wise manner and return the corresponding residuals matrix. This function is build on other functions (fit_models and residualsMatrix)

Usage

genSmoothCurveResiduals(cds, trend_formula = "~sm.ns(Pseudotime, df = 3)",
  relative_expr = T, residual_type = "response", cores = 1)

Arguments

Value

a data frame containing the data for the fitted spline curves.

Fit smooth spline curves and return the response matrix

Description

This function will fit smooth spline curves for the gene expression dynamics along pseudotime in a gene-wise manner and return the corresponding response matrix. This function is build on other functions (fit_models and responseMatrix) and used in calILRs and calABCs functions

Usage

genSmoothCurves(cds, new_data, trend_formula = "~sm.ns(Pseudotime, df = 3)",
  relative_expr = T, response_type = "response", cores = 1)

Arguments

Value

a data frame containing the data for the fitted spline curves.

Return the names of classic muscle genes

Description

Returns a list of classic muscle genes. Used to add conveinence for loading HSMM data.

Usage

get_classic_muscle_markers()

Import a seurat or scatter/scran CellDataSet object and convert it to a monocle cds.

Description

This function takes a monocle CellDataSet and converts it to another type of object used in another popular single cell analysis toolkit. It currently supports Scran and Seurat packages.

Usage

importCDS(otherCDS, import_all = FALSE)

Arguments

Value

a new monocle cell dataset object converted from other objects (Scatter or Seurat).

Examples

lung <- load_lung()
seurat_lung <- exportCDS(lung)
seurat_lung_all <- exportCDS(lung, export_all = T)
scater_lung <- exportCDS(lung, export_to = 'Scater')
scater_lung_all <- exportCDS(lung, export_to = 'Scater', export_all = T)

importCDS(seurat_lung)
importCDS(seurat_lung, import_all = T)
importCDS(seurat_lung_all)
importCDS(seurat_lung_all, import_all = T)
importCDS(scater_lung)
importCDS(scater_lung, import_all = T)
importCDS(scater_lung_all)
importCDS(scater_lung_all, import_all = T)

Build a CellDataSet from the HSMMSingleCell package

Description

Creates a cellDataSet using the data from the HSMMSingleCell package.

Usage

load_HSMM()

Return a CellDataSet of classic muscle genes.

Description

Return a CellDataSet of classic muscle genes.

Usage

load_HSMM_markers()

Value

A CellDataSet object

Build a CellDataSet from the data stored in inst/extdata directory.

Description

Build a CellDataSet from the data stored in inst/extdata directory.

Usage

load_lung()

Test genes for cell type-dependent expression

Description

takes a CellDataSet and a CellTypeHierarchy and classifies all cells into types passed functions passed into the CellTypeHierarchy. The function will remove all "Unknown" and "Ambiguous" types before identifying genes that are differentially expressed between types.

Usage

markerDiffTable(cds, cth, residualModelFormulaStr = "~1", balanced = FALSE,
  reclassify_cells = TRUE, remove_ambig = TRUE, remove_unknown = TRUE,
  verbose = FALSE, cores = 1)

Arguments

Value

A table of differential expression test results

Multicore apply-like function for CellDataSet

Description

mcesApply computes the row-wise or column-wise results of FUN, just like esApply. Variables in pData from X are available in FUN.

Usage

mcesApply(X, MARGIN, FUN, required_packages, cores = 1,
  convert_to_dense = TRUE, ...)

Arguments

Value

The result of with(pData(X) apply(exprs(X)), MARGIN, FUN, ...))

Retrieves the minimum spanning tree generated by Monocle during cell ordering.

Description

Retrieves the minimum spanning tree (MST) that Monocle constructs during orderCells(). This MST is mostly used in plot_spanning_tree to help assess the accuracy of Monocle's ordering.

Usage

minSpanningTree(cds)

Arguments

Value

An igraph object representing the CellDataSet's minimum spanning tree.

Examples

T <- minSpanningTree(HSMM)

Set the minimum spanning tree generated by Monocle during cell ordering.

Description

Sets the minimum spanning tree used by Monocle during cell ordering. Not intended to be called directly.

Usage

minSpanningTree(cds) <- value

Arguments

Value

An updated CellDataSet object

Examples

cds <- minSpanningTree(T)

Creates a new CellDateSet object.

Description

Creates a new CellDateSet object.

Usage

newCellDataSet(cellData, phenoData = NULL, featureData = NULL,
  lowerDetectionLimit = 0.1, expressionFamily = VGAM::negbinomial.size())

Arguments

Value

a new CellDataSet object

Examples

sample_sheet_small <- read.delim("../data/sample_sheet_small.txt", row.names=1)
sample_sheet_small$Time <- as.factor(sample_sheet_small$Time)
gene_annotations_small <- read.delim("../data/gene_annotations_small.txt", row.names=1)
fpkm_matrix_small <- read.delim("../data/fpkm_matrix_small.txt")
pd <- new("AnnotatedDataFrame", data = sample_sheet_small)
fd <- new("AnnotatedDataFrame", data = gene_annotations_small)
HSMM <- new("CellDataSet", exprs = as.matrix(fpkm_matrix_small), phenoData = pd, featureData = fd)

Classify cells according to a set of markers

Description

Creates a CellTypeHierarchy object which can store cell types with the addCellType() function. When classifyCells is used with a CellDataSet and a CellTypeHierarchy cells in the CellDataSet can be classified as cell types found in the CellTypeHierarchy

classifyCells accepts a cellDataSet and and a cellTypeHierarchy. Each cell in the cellDataSet is checked against the functions in the cellTypeHierarchy to determine each cell's type

Usage

newCellTypeHierarchy()
classifyCells(cds, cth, frequency_thresh = NULL, enrichment_thresh = NULL,
  ...)
calculateMarkerSpecificity(cds, cth, remove_ambig = TRUE,
  remove_unknown = TRUE)

Arguments

Details

CellTypeHierarchy objects are Monocle's mechanism for classifying cells into types based on known markers. To classify the cells in a CellDataSet object according to known markers, first construct a CellTypeHierachy with newCellTypeHierarchy() and addCellType() and then provide both the CellDataSet and the CellTypeHierachy to classifyCells() . Each call to addCellType() registers a classification function that accepts the expression data from a CellDataSet object as input, and returns a boolean vector indicating whether each cell is of the given type. When you call classifyCells() , each cell will be checked against the classification functions in the CellTypeHierachy . If you wish to make a cell type a subtype of another that's already been registered with a CellTypeHierarchy object, make that one the "parent" type with the cell_type_name argument. If you want two types to be mutually exclusive, make them "siblings" by giving them the same parent. The classifcation functions in a CellTypeHierarchy must take a single argument, a matrix of expression values, as input. Note that this matrix could either be a sparseMatrix or a dense matrix. Explicitly casting the input to a dense matrix inside a classification function is likely to drastically slow down classifyCells and other routines that use CellTypeHierarhcy objects. Successive calls to addCellType build up a tree of classification functions inside a CellTypeHierarchy. When two functions are siblings in the tree, classifyCells expects that a cell will meet the classification criteria for at most one of them. For example, you might place classification functions for T cells and B cells as siblings, because a cell cannot be both of these at the same time. When a cell meets the criteria for more than one function, it will be tagged as "Ambiguous". If classifyCells reports a large number of ambiguous cells, consider adjusting your classification functions. For example, some cells are defined by very high expression of a key gene that is expressed at lower levels in other cell types. Raising the threshold for this gene in a classification could resolve the ambiguities. A classification function can also have child functions. You can use this to specify subtypes of cells. For example, T cells express the gene CD3, and there are many subtypes. You can encode each subset by first adding a general T cell classification function that recognizes CD3, and then adding an additional function that recognizes CD4 (for CD4+ helper T cells), one for CD8 (to identify CD8+ cytotoxic T cells), and so on. classifyCells will aim to assign each cell to its most specific subtype in the "CellType" column. By default, classifyCells applies the classification functions to individual cells, but you can also apply it to cells in a "grouped" mode to impute the type of cells that are missing expression of your known markers. You can specify additional (quoted) grouping variables to classifyCells . The function will group the cells according to these factors, and then classify the cells. It will compute the frequency of each cell type in each group, and if a cell type is present at the frquency specified in frequency_thresh , all the cells in the group are classified as that type. If group contains more one cell type at this frequency, all the cells are marked "Ambiguous". This allows you to impute cell type based on unsupervised clustering results (e.g. with clusterCells ) or some other grouping criteria.

Value

newCellTypeHierarchy and addCellType both return an updated CellTypeHierarchy object. classifyCells returns an updated CellDataSet with a new column, "CellType", in the pData table.

For a CellDataset with N genes, and a CellTypeHierarchy with k types, returns a dataframe with N x k rows. Each row contains a gene and a specifity score for one of the types.

Examples

# Initialize a new CellTypeHierachy

# Register a set of classification functions. There are multiple types of T cells
# A cell cannot be both a B cell and a T cell, a T cell and a Monocyte, or
# a B cell and a Monocyte.
cth <- newCellTypeHierarchy()

cth <- addCellType(cth, "T cell",
classify_func=function(x) {x["CD3D",] > 0})

cth <- addCellType(cth, "CD4+ T cell",
classify_func=function(x) {x["CD4",] > 0},
parent_cell_type_name = "T cell")

cth <- addCellType(cth, "CD8+ T cell",
classify_func=function(x) {
|x["CD8A",] > 0 | x["CD8B",] > 0|
},
parent_cell_type_name = "T cell")

cth <- addCellType(cth, "B cell",
classify_func=function(x) {x["MS4A1",] > 0})

cth <- addCellType(cth, "Monocyte",
classify_func=function(x) {x["CD14",] > 0})

# Classify each cell in the CellDataSet "mix" according to these types
mix <- classifyCells(mix, cth)

# Group the cells by the pData table column "Cluster". Apply the classification
functions to the cells groupwise. If a group is at least 5% of a type, make
them all that type. If the group is 5% one type, and 5% a different, mutually
exclusive type, mark the whole cluster "Ambiguous"
mix <- classifyCells(mix, Cluster, 0.05)

Orders cells according to pseudotime.

Description

Learns a "trajectory" describing the biological process the cells are going through, and calculates where each cell falls within that trajectory. Monocle learns trajectories in two steps. The first step is reducing the dimensionality of the data with reduceDimension . The second is this function. function. This function takes as input a CellDataSet and returns it with two new columns: Pseudotime and State , which together encode where each cell maps to the trajectory. orderCells() optionally takes a "root" state, which you can use to specify the start of the trajectory. If you don't provide a root state, one is selected arbitrarily.

Usage

orderCells(cds, root_state = NULL, num_paths = NULL, reverse = NULL)

Arguments

Details

The reduction_method argument to reduceDimension determines which algorithm is used by orderCells() to learn the trajectory. If reduction_method == "ICA" , this function uses polygonal reconstruction to learn the underlying trajectory. If reduction_method == "DDRTree" , the trajectory is specified by the principal graph learned by the DDRTree function.

Whichever algorithm you use, the trajectory will be composed of segments. The cells from a segment will share the same value of State . One of these segments will be selected as the root of the trajectory arbitrarily. The most distal cell on that segment will be chosen as the "first" cell in the trajectory, and will have a Pseudotime value of zero. orderCells() will then "walk" along the trajectory, and as it encounters additional cells, it will assign them increasingly large values of Pseudotime.

Value

an updated CellDataSet object, in which phenoData contains values for State and Pseudotime for each cell

Return an ordering for a P node in the PQ tree

Description

Return an ordering for a P node in the PQ tree

Usage

order_p_node(q_level_list, dist_matrix)

Arguments

Plots the minimum spanning tree on cells. This function is deprecated.

Description

This function arranges all of the cells in the cds in a tree and predicts their location based on their pseudotime value

Usage

plot_spanning_tree(cds, x = 1, y = 2, color_by = "State",
  show_tree = TRUE, show_backbone = TRUE, backbone_color = "black",
  markers = NULL, show_cell_names = FALSE, cell_size = 1.5,
  cell_link_size = 0.75, cell_name_size = 2, show_branch_points = TRUE)

Arguments

Value

a ggplot2 plot object

Seealso

plot_cell_trajectory

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
plot_cell_trajectory(HSMM)
plot_cell_trajectory(HSMM, color_by="Pseudotime", show_backbone=FALSE)
plot_cell_trajectory(HSMM, markers="MYH3")

Plots clusters of cells .

Description

Plots clusters of cells .

Usage

plot_cell_clusters(cds, x = 1, y = 2, color_by = "Cluster",
  markers = NULL, show_cell_names = FALSE, cell_size = 1.5,
  cell_name_size = 2, ...)

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
HSMM <- reduceD
plot_cell_clusters(HSMM)
plot_cell_clusters(HSMM, color_by="Pseudotime")
plot_cell_clusters(HSMM, markers="MYH3")

Plots the minimum spanning tree on cells.

Description

Plots the minimum spanning tree on cells.

Usage

plot_cell_trajectory(cds, x = 1, y = 2, color_by = "State",
  show_tree = TRUE, show_backbone = TRUE, backbone_color = "black",
  markers = NULL, use_color_gradient = FALSE, markers_linear = FALSE,
  show_cell_names = FALSE, show_state_number = FALSE, cell_size = 1.5,
  cell_link_size = 0.75, cell_name_size = 2, state_number_size = 2.9,
  show_branch_points = TRUE, theta = 0, ...)

Arguments

Value

a ggplot2 plot object

Examples

lung <- load_lung()
plot_cell_trajectory(lung)
plot_cell_trajectory(lung, color_by="Pseudotime", show_backbone=FALSE)
plot_cell_trajectory(lung, markers="MYH3")

Plots kinetic clusters of genes.

Description

returns a ggplot2 object showing the shapes of the expression patterns followed by a set of pre-selected genes. The topographic lines highlight the distributions of the kinetic patterns relative to overall trend lines.

Usage

plot_clusters(cds, clustering, drawSummary = TRUE, sumFun = mean_cl_boot,
  ncol = NULL, nrow = NULL, row_samples = NULL, callout_ids = NULL)

Arguments

Value

a ggplot2 plot object

Examples

full_model_fits <- fitModel(HSMM_filtered[sample(nrow(fData(HSMM_filtered)), 100),],
modelFormulaStr="~VGAM::bs(Pseudotime)")
expression_curve_matrix <- responseMatrix(full_model_fits)
clusters <- clusterGenes(expression_curve_matrix, k=4)
plot_clusters(HSMM_filtered[ordering_genes,], clusters)

Not sure we're ready to release this one quite yet: Plot the branch genes in pseduotime with separate branch curves

Description

Not sure we're ready to release this one quite yet: Plot the branch genes in pseduotime with separate branch curves

Usage

plot_coexpression_matrix(cds, rowgenes, colgenes, relative_expr = TRUE,
  min_expr = NULL, cell_size = 0.85, label_by_short_name = TRUE,
  show_density = TRUE, round_expr = FALSE)

Arguments

Value

a ggplot2 plot object

Plots the minimum spanning tree on cells.

Description

Plots the minimum spanning tree on cells.

Usage

plot_complex_cell_trajectory(cds, x = 1, y = 2, root_states = NULL,
  color_by = "State", show_tree = TRUE, show_backbone = TRUE,
  backbone_color = "black", markers = NULL, show_cell_names = FALSE,
  cell_size = 1.5, cell_link_size = 0.75, cell_name_size = 2,
  show_branch_points = TRUE, ...)

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
plot_complex_cell_trajectory(HSMM)
plot_complex_cell_trajectory(HSMM, color_by="Pseudotime", show_backbone=FALSE)
plot_complex_cell_trajectory(HSMM, markers="MYH3")

Create a heatmap to demonstrate the bifurcation of gene expression along two branchs

@description returns a heatmap that shows changes in both lineages at the same time. It also requires that you choose a branch point to inspect. Columns are points in pseudotime, rows are genes, and the beginning of pseudotime is in the middle of the heatmap. As you read from the middle of the heatmap to the right, you are following one lineage through pseudotime. As you read left, the other. The genes are clustered hierarchically, so you can visualize modules of genes that have similar lineage-dependent expression patterns.

Description

Create a heatmap to demonstrate the bifurcation of gene expression along two branchs

@description returns a heatmap that shows changes in both lineages at the same time. It also requires that you choose a branch point to inspect. Columns are points in pseudotime, rows are genes, and the beginning of pseudotime is in the middle of the heatmap. As you read from the middle of the heatmap to the right, you are following one lineage through pseudotime. As you read left, the other. The genes are clustered hierarchically, so you can visualize modules of genes that have similar lineage-dependent expression patterns.

Usage

plot_genes_branched_heatmap(cds_subset, branch_point = 1,
  branch_states = NULL, branch_labels = c("Cell fate 1", "Cell fate 2"),
  cluster_rows = TRUE, hclust_method = "ward.D2", num_clusters = 6,
  hmcols = NULL, branch_colors = c("#979797", "#F05662", "#7990C8"),
  add_annotation_row = NULL, add_annotation_col = NULL,
  show_rownames = FALSE, use_gene_short_name = TRUE, scale_max = 3,
  scale_min = -3, norm_method = c("log", "vstExprs"),
  trend_formula = "~sm.ns(Pseudotime, df=3) * Branch",
  return_heatmap = FALSE, cores = 1, ...)

Arguments

Value

A list of heatmap_matrix (expression matrix for the branch committment), ph (pheatmap heatmap object), annotation_row (annotation data.frame for the row), annotation_col (annotation data.frame for the column).

Plot the branch genes in pseduotime with separate branch curves.

Description

Works similarly to plot_genes_in_psuedotime esceptit shows one kinetic trend for each lineage.

Usage

plot_genes_branched_pseudotime(cds, branch_states = NULL, branch_point = 1,
  branch_labels = NULL, method = "fitting", min_expr = NULL,
  cell_size = 0.75, nrow = NULL, ncol = 1, panel_order = NULL,
  color_by = "State", expression_curve_linetype_by = "Branch",
  trend_formula = "~ sm.ns(Pseudotime, df=3) * Branch",
  reducedModelFormulaStr = NULL, label_by_short_name = TRUE,
  relative_expr = TRUE, ...)

Arguments

Details

This plotting function is used to make the branching plots for a branch dependent gene goes through the progenitor state and bifurcating into two distinct branchs (Similar to the pitch-fork bifurcation in dynamic systems). In order to make the bifurcation plot, we first duplicated the progenitor states and by default stretch each branch into maturation level 0-100. Then we fit two nature spline curves for each branchs using VGAM package.

Value

a ggplot2 plot object

Plots expression for one or more genes as a function of pseudotime

Description

Plots expression for one or more genes as a function of pseudotime. Plotting allows you determine if the ordering produced by orderCells() is correct and it does not need to be flipped using the "reverse" flag in orderCells

Usage

plot_genes_in_pseudotime(cds_subset, min_expr = NULL, cell_size = 0.75,
  nrow = NULL, ncol = 1, panel_order = NULL, color_by = "State",
  trend_formula = "~ sm.ns(Pseudotime, df=3)", label_by_short_name = TRUE,
  relative_expr = TRUE, vertical_jitter = NULL, horizontal_jitter = NULL)

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
my_genes <- row.names(subset(fData(HSMM), gene_short_name %in% c("CDK1", "MEF2C", "MYH3")))
cds_subset <- HSMM[my_genes,]
plot_genes_in_pseudotime(cds_subset, color_by="Time")

Plots expression for one or more genes as a jittered, grouped points

Description

Accepts a subset of a CellDataSet and an attribute to group cells by, and produces one or more ggplot2 objects that plots the level of expression for each group of cells.

Usage

plot_genes_jitter(cds_subset, grouping = "State", min_expr = NULL,
  cell_size = 0.75, nrow = NULL, ncol = 1, panel_order = NULL,
  color_by = NULL, plot_trend = FALSE, label_by_short_name = TRUE,
  relative_expr = TRUE)

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
my_genes <- HSMM[row.names(subset(fData(HSMM), gene_short_name %in% c("MYOG", "ID1", "CCNB2"))),]
plot_genes_jitter(my_genes, grouping="Media", ncol=2)

Plots the number of cells expressing one or more genes as a barplot

Description

@description Accetps a CellDataSet and a parameter,"grouping", used for dividing cells into groups. Returns one or more bar graphs (one graph for each gene in the CellDataSet). Each graph shows the percentage of cells that express a gene in the in the CellDataSet for each sub-group of cells created by "grouping".

Let's say the CellDataSet passed in included genes A, B, and C and the "grouping parameter divided all of the cells into three groups called X, Y, and Z. Then three graphs would be produced called A, B, and C. In the A graph there would be three bars one for X, one for Y, and one for Z. So X bar in the A graph would show the percentage of cells in the X group that express gene A.

Usage

plot_genes_positive_cells(cds_subset, grouping = "State", min_expr = 0.1,
  nrow = NULL, ncol = 1, panel_order = NULL, plot_as_fraction = TRUE,
  label_by_short_name = TRUE, relative_expr = TRUE, plot_limits = c(0,
  100))

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
MYOG_ID1 <- HSMM[row.names(subset(fData(HSMM), gene_short_name %in% c("MYOG", "ID1"))),]
plot_genes_positive_cells(MYOG_ID1, grouping="Media", ncol=2)

Plots expression for one or more genes as a violin plot

Description

Accepts a subset of a CellDataSet and an attribute to group cells by, and produces one or more ggplot2 objects that plots the level of expression for each group of cells.

Usage

plot_genes_violin(cds_subset, grouping = "State", min_expr = NULL,
  cell_size = 0.75, nrow = NULL, ncol = 1, panel_order = NULL,
  color_by = NULL, plot_trend = FALSE, label_by_short_name = TRUE,
  relative_expr = TRUE, log_scale = TRUE)

Arguments

Value

a ggplot2 plot object

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
my_genes <- HSMM[row.names(subset(fData(HSMM), gene_short_name %in% c("ACTA1", "ID1", "CCNB2"))),]
plot_genes_violin(my_genes, grouping="Hours", ncol=2, min_expr=0.1)

Create a heatmap to demonstrate the bifurcation of gene expression along multiple branches

Description

Create a heatmap to demonstrate the bifurcation of gene expression along multiple branches

Usage

plot_multiple_branches_heatmap(cds, branches, branches_name = NULL,
  cluster_rows = TRUE, hclust_method = "ward.D2", num_clusters = 6,
  hmcols = NULL, add_annotation_row = NULL, add_annotation_col = NULL,
  show_rownames = FALSE, use_gene_short_name = TRUE,
  norm_method = c("vstExprs", "log"), scale_max = 3, scale_min = -3,
  trend_formula = "~sm.ns(Pseudotime, df=3)", return_heatmap = FALSE,
  cores = 1)

Arguments

Value

A list of heatmap_matrix (expression matrix for the branch committment), ph (pheatmap heatmap object), annotation_row (annotation data.frame for the row), annotation_col (annotation data.frame for the column).

Create a kinetic curves to demonstrate the bifurcation of gene expression along multiple branches

Description

Create a kinetic curves to demonstrate the bifurcation of gene expression along multiple branches

Usage

plot_multiple_branches_pseudotime(cds, branches, branches_name = NULL,
  min_expr = NULL, cell_size = 0.75, norm_method = c("vstExprs", "log"),
  nrow = NULL, ncol = 1, panel_order = NULL, color_by = "Branch",
  trend_formula = "~sm.ns(Pseudotime, df=3)", label_by_short_name = TRUE,
  TPM = FALSE, cores = 1)

Arguments

Value

a ggplot2 plot object

Plots genes by mean vs. dispersion, highlighting those selected for ordering

Description

Each gray point in the plot is a gene. The black dots are those that were included in the last call to setOrderingFilter. The red curve shows the mean-variance model learning by estimateDispersions().

Usage

plot_ordering_genes(cds)

Arguments

Plots the percentage of variance explained by the each component based on PCA from the normalized expression data using the same procedure used in reduceDimension function.

Description

Plots the percentage of variance explained by the each component based on PCA from the normalized expression data using the same procedure used in reduceDimension function.

Usage

plot_pc_variance_explained(cds, max_components = 100, norm_method = c("log",
  "vstExprs", "none"), residualModelFormulaStr = NULL, pseudo_expr = NULL,
  return_all = F, use_existing_pc_variance = FALSE, verbose = FALSE, ...)

Arguments

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
plot_pc_variance_explained(HSMM)

Plots a pseudotime-ordered, row-centered heatmap

Description

The function plot_pseudotime_heatmap takes a CellDataSet object (usually containing a only subset of significant genes) and generates smooth expression curves much like plot_genes_in_pseudotime. Then, it clusters these genes and plots them using the pheatmap package. This allows you to visualize modules of genes that co-vary across pseudotime.

Usage

plot_pseudotime_heatmap(cds_subset, cluster_rows = TRUE,
  hclust_method = "ward.D2", num_clusters = 6, hmcols = NULL,
  add_annotation_row = NULL, add_annotation_col = NULL,
  show_rownames = FALSE, use_gene_short_name = TRUE,
  norm_method = c("log", "vstExprs"), scale_max = 3, scale_min = -3,
  trend_formula = "~sm.ns(Pseudotime, df=3)", return_heatmap = FALSE,
  cores = 1)

Arguments

Value

A list of heatmap_matrix (expression matrix for the branch committment), ph (pheatmap heatmap object), annotation_row (annotation data.frame for the row), annotation_col (annotation data.frame for the column).

Plots the decision map of density clusters .

Description

Plots the decision map of density clusters .

Usage

plot_rho_delta(cds, rho_threshold = NULL, delta_threshold = NULL)

Arguments

Examples

library(HSMMSingleCell)
HSMM <- load_HSMM()
plot_rho_delta(HSMM)

Recursively builds and returns a PQ tree for the MST

Description

Recursively builds and returns a PQ tree for the MST

Usage

pq_helper(mst, use_weights = TRUE, root_node = NULL)

Arguments

Compute a projection of a CellDataSet object into a lower dimensional space

Description

Monocle aims to learn how cells transition through a biological program of gene expression changes in an experiment. Each cell can be viewed as a point in a high-dimensional space, where each dimension describes the expression of a different gene in the genome. Identifying the program of gene expression changes is equivalent to learning a trajectory that the cells follow through this space. However, the more dimensions there are in the analysis, the harder the trajectory is to learn. Fortunately, many genes typically co-vary with one another, and so the dimensionality of the data can be reduced with a wide variety of different algorithms. Monocle provides two different algorithms for dimensionality reduction via reduceDimension . Both take a CellDataSet object and a number of dimensions allowed for the reduced space. You can also provide a model formula indicating some variables (e.g. batch ID or other technical factors) to "subtract" from the data so it doesn't contribute to the trajectory.

Usage

reduceDimension(cds, max_components = 2, reduction_method = c("DDRTree",
  "ICA", "tSNE", "SimplePPT", "L1-graph", "SGL-tree"), norm_method = c("log",
  "vstExprs", "none"), residualModelFormulaStr = NULL, pseudo_expr = 1,
  relative_expr = TRUE, auto_param_selection = TRUE, verbose = FALSE,
  scaling = TRUE, ...)

Arguments

Details

You can choose two different reduction algorithms: Independent Component Analysis (ICA) and Discriminative Dimensionality Reduction with Trees (DDRTree). The choice impacts numerous downstream analysis steps, including orderCells . Choosing ICA will execute the ordering procedure described in Trapnell and Cacchiarelli et al., which was implemented in Monocle version 1. DDRTree is a more recent manifold learning algorithm developed by Qi Mao and colleages. It is substantially more powerful, accurate, and robust for single-cell trajectory analysis than ICA, and is now the default method.

Often, experiments include cells from different batches or treatments. You can reduce the effects of these treatments by transforming the data with a linear model prior to dimensionality reduction. To do so, provide a model formula through residualModelFormulaStr .

Prior to reducing the dimensionality of the data, it usually helps to normalize it so that highly expressed or highly variable genes don't dominate the computation. reduceDimension() automatically transforms the data in one of several ways depending on the expressionFamily of the CellDataSet object. If the expressionFamily is negbinomial or negbinomial.size , the data are variance-stabilized. If the expressionFamily is Tobit , the data are adjusted by adding a pseudocount (of 1 by default) and then log-transformed. If you don't want any transformation at all, set norm_method to "none" and pseudo_expr to 0. This maybe useful for single-cell qPCR data, or data you've already transformed yourself in some way.

Value

an updated CellDataSet object

Get the weights needed to lift cells back to high dimensional expression space.

Description

Retrieves the weights that transform the cells' coordinates in the reduced dimension space back to the full (whitened) space.

Usage

reducedDimA(cds)

Arguments

Value

A matrix that when multiplied by a reduced-dimension set of coordinates for the CellDataSet, recovers a matrix in the full (whitened) space

Examples

A <- reducedDimA(HSMM)

Get the weights needed to lift cells back to high dimensional expression space.

Description

Sets the weights transform the cells' coordinates in the reduced dimension space back to the full (whitened) space.

Usage

reducedDimA(cds) <- value

Arguments

Value

An updated CellDataSet object

Examples

cds <- reducedDimA(A)

Retrieves the the whitening matrix during independent component analysis.

Description

Retrieves the the whitening matrix during independent component analysis.

Usage

reducedDimK(cds)

Arguments

Value

A matrix, where each row is a set of whitened expression values for a feature and columns are cells.

Examples

K <- reducedDimW(HSMM)

Sets the the whitening matrix during independent component analysis.

Description

Sets the the whitening matrix during independent component analysis.

Usage

reducedDimK(cds) <- value

Arguments

Value

A matrix, where each row is a set of whitened expression values for a feature and columns are cells.

Examples

cds <- reducedDimK(K)

Retrieves the coordinates of each cell in the reduced-dimensionality space generated by calls to reduceDimension.

Description

Reducing the dimensionality of the expression data is a core step in the Monocle workflow. After you call reduceDimension(), this function will return the new coordinates of your cells in the reduced space.

Usage

reducedDimS(cds)

Arguments

Value

A matrix, where rows are cell coordinates and columns correspond to dimensions of the reduced space.

Examples

S <- reducedDimS(HSMM)

Set embedding coordinates of each cell in a CellDataSet.

Description

This function sets the coordinates of each cell in a new (reduced-dimensionality) space. Not intended to be called directly.

Usage

reducedDimS(cds) <- value

Arguments

Value

An update CellDataSet object

Examples

cds <- reducedDimS(S)

Get the whitened expression values for a CellDataSet.

Description

Retrieves the expression values for each cell (as a matrix) after whitening during dimensionality reduction.

Usage

reducedDimW(cds)

Arguments

Value

A matrix, where each row is a set of whitened expression values for a feature and columns are cells.

Examples

W <- reducedDimW(HSMM)

Sets the whitened expression values for each cell prior to independent component analysis. Not intended to be called directly.

Description

Sets the whitened expression values for each cell prior to independent component analysis. Not intended to be called directly.

Usage

reducedDimW(cds) <- value

Arguments

Value

An updated CellDataSet object

Examples

#' cds <- reducedDimA(A)

Transform relative expression values into absolute transcript counts.

Description

Converts FPKM/TPM data to transcript counts. This allows for the use for negative binomial as an expressionFamily. These results are often far more accurate than using tobit().

Usage

relative2abs(relative_cds, t_estimate = estimate_t(exprs(relative_cds)),
  modelFormulaStr = "~1", ERCC_controls = NULL, ERCC_annotation = NULL,
  volume = 10, dilution = 40000, mixture_type = 1,
  detection_threshold = 800, expected_capture_rate = 0.25,
  verbose = FALSE, return_all = FALSE, method = c("num_genes",
  "tpm_fraction"), cores = 1)

Arguments

Details

Transform a relative expression matrix to absolute transcript matrix based on the inferred linear regression parameters from most abundant isoform relative expression value. This function takes a relative expression matrix and a vector of estimated most abundant expression value from the isoform-level matrix and transform it into absolute transcript number. It is based on the observation that the recovery efficient of the single-cell RNA-seq is relative low and that most expressed isoforms of gene in a single cell therefore only sequenced one copy so that the most abundant isoform log10-FPKM (t^) will corresponding to 1 copy transcript. It is also based on the fact that the spikein regression parameters k/b for each cell will fall on a line because of the intrinsic properties of spikein experiments. We also assume that if we perform the same spikein experiments as Treutlein et al. did, the regression parameters should also fall on a line in the same way. The function takes the the vector t^ and the detection limit as input, then it uses the t^ and the m/c value corresponding to the detection limit to calculate two parameters vectors k^ and b^ (corresponding to each cell) which correspond to the slope and intercept for the linear conversion function between log10 FPKM and log10 transcript counts. The function will then apply a linear transformation to convert the FPKM to estimated absolute transcript counts based on the the k^ and b^*. The default m/c values used in the algoritm are 3.652201, 2.263576, respectively.

Value

an matrix of absolute count for isoforms or genes after the transformation.

Examples

HSMM_relative_expr_matrix <- exprs(HSMM)
HSMM_abs_matrix <- relative2abs(HSMM_relative_expr_matrix,
t_estimate = estimate_t(HSMM_relative_expr_matrix))

Response values

Description

Generates a matrix of response values for a set of fitted models

Usage

residualMatrix(models, residual_type = "response", cores = 1)

Arguments

Value

a matrix where each row is a vector of response values for a particular feature's model, and columns are cells.

Calculates response values.

Description

Generates a matrix of response values for a set of fitted models

Usage

responseMatrix(models, newdata = NULL, response_type = "response",
  cores = 1)

Arguments

Value

a matrix where each row is a vector of response values for a particular feature's model, and columns are cells.

Select the most cell type specific markers

Description

This is a handy wrapper function around dplyr's top_n function to extract the most specific genes for each cell type. Convenient, for example, for selecting a balanced set of genes to be used in semi-supervised clustering or ordering.

Usage

selectTopMarkers(marker_specificities, num_markers = 10)

Arguments

Value

A data frame of specificity results

Marks genes for clustering

Description

The function marks genes that will be used for clustering in subsequent calls to clusterCells. The list of selected genes can be altered at any time.

Usage

setOrderingFilter(cds, ordering_genes)

Arguments

Value

an updated CellDataSet object

Spike-in transcripts data.

Description

A dataset containing the information for the 92 ERCC spikein transcripts (This dataset is based on the data from the Nature paper from Stephen Quake group)

Format

A data frame with 92 rows and 9 variables: list(" ", " ", list(list("ERCC_ID"), list("ID for ERCC transcripts")), " ", " ", list(list("subgroup"), list("Subgroup for ERCC transcript")), " ", " ", list(list("conc_attomoles_ul_Mix1"), list("Contration of Mix 1 (attomoles / ul)")), " ", " ", list(list("conc_attomoles_ul_Mix2"), list("Contration of Mix 2 (attomoles / ul)")), " ", " ", list(list("exp_fch_ratio"), list("expected fold change between mix 1 over mix 2")), " ", " ", list(list("numMolecules"), list("number of molecules calculated from concentration and volume")),

"

", " ", list(list("rounded_numMolecules"), list("number in rounded digit of molecules calculated from concentration and volume")), " ")

Usage

spike_df

Return a variance-stabilized matrix of expression values

Description

This function was taken from the DESeq package (Anders and Huber) and modified to suit Monocle's needs. It accpets a either a CellDataSet or the expression values of one and returns a variance-stabilized matrix based off of them.

Usage

vstExprs(cds, dispModelName = "blind", expr_matrix = NULL,
  round_vals = TRUE)

Arguments