SPEAR: SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

Plotting results from a SPEARobject

Required Libraries:

library(SPEAR)

Overview:

This vignette will highlight the main plotting functions in the SPEAR package.

Plots depend on the family parameter (gaussian, binomial, ordinal, or multinomial).

Current Plotting Functions:

plot.factor.scores
plot.embedding
plot.contributions
plot.feature.expression
plot.features
plot.cv.loss
plot.predictions
plot.circos
plot.network
plot.coefficients
plot.probabilities

`plot.factor.scores`

Plot factor scores from a SPEARobject for a particular dataset data.

Parameters:

data - Which dataset to use? Can be any dataset listed under SPEARobj$data$____. Defaults to "train".
cv - If data = "train", get factor scores generated from $run.cv.spear(...)? If $run.spear(...) was used or if data != "train" this parameter is ignored. Defaults to FALSE. NOTE: There is an element of randomness if the factor scores are not correlated with the response, so it is recommended to view the factor scores with cv = FALSE
group - What to plot on the x.axis? Defaults to “response” (indicated by SPEARobj$options$current.response.idx), but can be any feature name found in colnames(SPEARobj$data$train$X)
factors Which factors to show? Defaults to NULL (all factors). Use a vector of integers (i.e. c(1, 2, 4)) to specify fewer factors.
show.legend Show the legend of the plot (for group)? Defaults to TRUE

Examples:

# Default Parameters:
# Plot CV factor scores vs. Response:
SPEARobj$plot.factor.scores()

# Use a feature name found in colnames(SPEARobj$data$train$X):
feature.name <- "OmicsData4_feat194"
SPEARobj$plot.factor.scores(group = feature.name)

`plot.embedding`

Plot factor scores on x and y axis from a SPEARobject for a particular dataset data.

Parameters:

data - Which dataset to use? Can be any dataset listed under SPEARobj$data$____. Defaults to "train".
cv - If data = "train", get factor scores generated from $run.cv.spear(...)? If $run.spear(...) was used or if data != "train" this parameter is ignored. Defaults to FALSE. NOTE: There is an element of randomness if the factor scores are not correlated with the response, so it is recommended to view the factor scores with cv = FALSE
group - How to color points? Defaults to “response” (indicated by SPEARobj$options$current.response.idx), but can be any feature name found in colnames(SPEARobj$data$train$X)
x Which factor to show on the x axis? Defaults to 1.
y Which factor to show on the y axis? Defaults to 2.
show.legend Show the legend of the plot (for group)? Defaults to TRUE

Examples:

# Default Parameters:
# Factor 1 vs. Factor 2, colored by response
SPEARobj$plot.embedding()

# Different factors
SPEARobj$plot.embedding(x = 3, y = 4)

# Color by a feature:
# Get the top contributing feature for Factor 1:
feature.name <- SPEARobj$get.features(factors = 1)$Feature[1]
# Color by the top contributing feature ('OmicsData4_feat194')
SPEARobj$plot.embedding(group = "OmicsData4_feat194")

`plot.contributions`

Plot factor contributions for a SPEARobject.

Parameters:

do.X Calculate contributions for X? Defaults to TRUE
do.Y Calculate contributions for Y? Defaults to TRUE

Examples:

SPEARobj$plot.contributions()

SPEARobj$plot.contributions(do.X = FALSE)

`plot.feature.expression`

Plot features from a SPEARobject.

Parameters:

factor Which factor to return features for? Accepts a single integer or factors = 1) Defaults to 1 (factor 1).
datasets Which datasets? Check names via names(SPEARobject$data$train$Xlist). Defaults to NULL (all)
order.by.factor.score Order the samples by factor score? Defaults to TRUE
order.by.sign Order the samples by projection.coefficient sign? Defaults to TRUE
show.factor.scores Should the factor scores be plotted on top of the heatmap? Defaults to TRUE
data Which data? Defaults to "train".
cv Use CV factor scores
rank How to rank features? Defaults to "probability" (joint.probability) followed by projection.coefficient. Can also be +cent” (only projection.coefficient magnitude) .
coefficient.cutoff What projection.coefficient value to use as the cutoff? Defaults to 0
probability.cutoff What joint.probability value to use as the cutoff? Defaults to .95
plot.by.dataset Should the features be separated by dataset? Defaults to TRUE
show.top.n Stop showing features after the top stop.by.n features. Defaults to 50. Set to a number higher than the total features to plot all of them .
show.all.names Should all feature names be shown? Defaults to FALSE (if num.features > 100, can be very crowded).
abbrev.names Abbreviate long feature names to 20 characters? Defaults to TRUE.
dark Use dark theme? Defaults to TRUE

# Defaults to Factor 1:
SPEARobj$plot.feature.expression()

# Show different factor:
# Set different probability cutoff:
SPEARobj$plot.feature.expression(factor = 2, probability.cutoff = .975)

# Hide the factor scores:
SPEARobj$plot.feature.expression(show.factor.scores = FALSE)

# Disable the dark color scheme
SPEARobj$plot.feature.expression(dark = FALSE)

`plot.features`

Plot features from a SPEARobject.

Parameters:

rank How to rank features? Defaults to "probability" (joint.probability) followed by projection.coefficient. Can also be “coefficient” (only projection.coefficient magnitude).
factor Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL (all).
datasets Which datasets? Check names via names(SPEARobject$data$train$Xlist). Defaults to NULL (all)
coefficient.cutoff What projection.coefficient value to use as the cutoff? Defaults to 0
probability.cutoff What joint.probability value to use as the cutoff? Defaults to .95
plot.by.dataset Should the features be separated by dataset? Defaults to TRUE
show.top.n Stop showing features after the top stop.by.n features. Defaults to 100. Set to a number higher than the total number of features to plot all of them.
abbrev.names Abbreviate long feature names to 20 characters? Defaults to TRUE.

Examples:

# Default: Get top features > .95 joint.probability for Factor 1 (for current w.idx model):
SPEARobj$plot.features()

# Loosen the cutoffs: (set probability.cutoff to .5 instead of .95, default)
SPEARobj$plot.features(probability.cutoff = .75)

SPEARobj$plot.features(factor = 4)

# Plot features by their order, ignoring dataset grouping:
SPEARobj$plot.features(plot.by.dataset = FALSE)

`plot.cv.loss`

Plot the mean cross validated error for a SPEAR model trained with run.cv.spear() and cv.evaluate().

Parameters:

show.labels Show the mean cv error labels? Defaults to TRUE

Examples:

SPEARobj$plot.cv.loss()

SPEARobj$plot.cv.loss(show.labels = FALSE)

`plot.factor.correlations`

Plot factor score correlations for various SPEAR models (w.idxs)

Parameters:

factors Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL
w.idxs Which w.idxs (models) to use? Accepts integers (i.e.w.idxs = c(1,2,4)orw.idxs = 1) Defaults toNULL` (all).
data Which dataset to use? Can be any dataset listed under $data$____. Defaults to "train".
cv If data = "train", get factor scores generated from $run.cv.spear? If $run.spear was used or if data != "train" this is ignored. Defaults to TRUE.
method Which correlation method to use? Defaults to spearman. Accepts spearman, pearson, and kendall.
label How to label/distinguish factors per model? Defaults to x (label with w.x). Accepts x, y, xy, idx, all.
sig.cutoff When to plot an asterisk (*)? Will plot if p.value from cor.test is <= sig.cutoff. Defaults to .05

Examples:

SPEARobj$plot.factor.correlations()

# Choose specific SPEAR models (via w.idx)
SPEARobj$plot.factor.correlations(w.idxs = c(1, 2, 3))

# Choose specific factors:
SPEARobj$plot.factor.correlations(factors = c(1, 2))

# Choose specific factors:
SPEARobj$plot.factor.correlations(factors = c(1, 2))

`plot.predictions`

Plot predictions from a SPEARobject and a dataset data.

Parameters:

data Which dataset to use? Can be any dataset listed under $data$____. Defaults to "train".
cv If data = "train", get factor scores generated from $run.cv.spear? If $run.spear was used or if data != "train" this parameter is ignored. Defaults to TRUE.
color.by.error Should gaussian data be colored by squared residual error? Defaults to TRUE

Examples:

SPEARobj$plot.predictions()

SPEARobj$plot.predictions(color.by.error = FALSE)

`plot.circos`

Plot a circos plot (chord diagram) for a correlation matrix supplied. Rows and columns need to be found in colnames(SPEARobj$data$train$X).

Parameters:

correlation.matrix Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL (all).
correlation.cutoff What is the correlation cutoff to be included in the plot? Defaults to 0.5
allow.same.dataset Should connections within the same dataset be plotted? Defaults to FALSE
sign Which direction of correlations to show? Can be "both" (default), "pos" (positive), or "negative" (negative).
highlight.feature Should one feature be highlighted in the plot? Can be a string (the feature name). Feature must be found in the final plot to be highlighted. Defaults to NULL (no highlighting)
highlight.color What color to highlight the highlight.feature connections? Defaults to "black"
highlight.emphasis How much should the highlight.feature connections be emphasized? 1 will put the same size as other connections. Defaults to 2.5

Examples:

# If no correlation.matrix argument is supplied, will automatically make
#    one for Factor 1 with SPEARobj$get.correlation.matrix()
SPEARobj$plot.circos(correlation.cutoff = .3)

## No argument passed for correlation.matrix, getting correlation matrix default (for Factor 1)...
## Getting correlation matrix for Factor 1
## Found 32 features for Factor 1
## Beginning pearson correlation tests for 32 features
## 0----------50--------100%
##  ====================
## Plotting the following correlations with magnitude > 0.3:
## - OmicsData1 | 0
## - OmicsData2 | 0
## - OmicsData3 | 11
## - OmicsData4 | 6

## Warning in par(c(0, 0, 0, 0)): argument 1 does not name a graphical parameter

# Use the get.correlation.matrix function to get a correlation matrix:
correlation.matrix <- SPEARobj$get.correlation.matrix(factor = 4)

## Getting correlation matrix for Factor 4
## Found 34 features for Factor 4
## Beginning pearson correlation tests for 34 features
## 0----------50--------100%
##  ======================

SPEARobj$plot.circos(correlation.matrix, correlation.cutoff = .1)

## Plotting the following correlations with magnitude > 0.1:
## - OmicsData1 | 15
## - OmicsData2 | 19
## - OmicsData3 | 0
## - OmicsData4 | 0

## Warning in par(c(0, 0, 0, 0)): argument 1 does not name a graphical parameter

# Highlight a particular feature:
SPEARobj$plot.circos(correlation.matrix, 
                     correlation.cutoff = .1,
                     highlight.feature = "OmicsData2_feat192",
                     highlight.color = "black")

## Plotting the following correlations with magnitude > 0.1:
## - OmicsData1 | 15
## - OmicsData2 | 19
## - OmicsData3 | 0
## - OmicsData4 | 0

## Warning in par(c(0, 0, 0, 0)): argument 1 does not name a graphical parameter

# Allow features from the same dataset to connect to one another
SPEARobj$plot.circos(correlation.matrix,
                     correlation.cutoff = .3,
                     allow.same.dataset = TRUE)

## Plotting the following correlations with magnitude > 0.3:
## - OmicsData1 | 3
## - OmicsData2 | 2
## - OmicsData3 | 0
## - OmicsData4 | 0

## Warning in par(c(0, 0, 0, 0)): argument 1 does not name a graphical parameter

`plot.network`

Plot a network (using the visNetwork package) from a correlation matrix supplied. Rows and columns need to be found in colnames(SPEARobj$data$train$X).

Parameters:

correlation.matrix Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL (all).
correlation.cutoff What is the correlation cutoff to be included in the plot? Defaults to 0.5
allow.same.dataset Should connections within the same dataset be plotted? Defaults to FALSE
sign Which direction of correlations to show? Can be "both" (default), "pos" (positive), or "negative" (negative).

Examples:

# If no correlation.matrix argument is supplied, will automatically make
#    one for Factor 1 with SPEARobj$get.correlation.matrix()
SPEARobj$plot.network()

## No argument passed for correlation.matrix, getting correlation matrix default (for Factor 1)...
## Getting correlation matrix for Factor 1
## Found 32 features for Factor 1
## Beginning pearson correlation tests for 32 features
## 0----------50--------100%
##  ====================

{"x":{"nodes":{"id":["OmicsData3_feat21","OmicsData3_feat78","OmicsData4_feat194"],"label":["OmicsData3_feat21","OmicsData3_feat78","OmicsData4_feat194"],"color":["#DB5B68","#DB5B68","#FAA138"],"Dataset":["OmicsData3","OmicsData3","OmicsData4"],"shadow":[true,true,true],"borderWidth":[1,1,1],"value":[1,1,2],"x":[-1,1,-0.00365929106042595],"y":[-0.590077360199939,-1,1]},"edges":{"from":["OmicsData3_feat21","OmicsData3_feat78"],"to":["OmicsData4_feat194","OmicsData4_feat194"],"width":[-2.66150985311116,2.42701521898015],"color":["blue","red"],"smooth":[false,false],"shadow":[false,false]},"nodesToDataframe":true,"edgesToDataframe":true,"options":{"width":"100%","height":"100%","nodes":{"shape":"dot","physics":false},"manipulation":{"enabled":false},"edges":{"smooth":false},"physics":{"stabilization":false}},"groups":null,"width":null,"height":null,"idselection":{"enabled":true,"style":"width: 150px; height: 26px","useLabels":true,"main":"Select by id"},"byselection":{"enabled":true,"style":"width: 150px; height: 26px","multiple":false,"hideColor":"rgba(200,200,200,0.5)","highlight":false,"variable":"Dataset","main":"Select by Dataset","values":["OmicsData3","OmicsData4"]},"main":null,"submain":null,"footer":null,"background":"rgba(0, 0, 0, 0)","highlight":{"enabled":true,"hoverNearest":false,"degree":1,"algorithm":"all","hideColor":"rgba(200,200,200,0.5)","labelOnly":true},"collapse":{"enabled":false,"fit":false,"resetHighlight":true,"clusterOptions":null,"keepCoord":true,"labelSuffix":"(cluster)"},"igraphlayout":{"type":"square"}},"evals":[],"jsHooks":[]}

# Use the get.correlation.matrix function to get a correlation matrix:
correlation.matrix <- SPEARobj$get.correlation.matrix(factor = 4)

## Getting correlation matrix for Factor 4
## Found 34 features for Factor 4
## Beginning pearson correlation tests for 34 features
## 0----------50--------100%
##  ======================

SPEARobj$plot.network(correlation.matrix, correlation.cutoff = .2)

{"x":{"nodes":{"id":["OmicsData1_feat95","OmicsData2_feat105","OmicsData2_feat112","OmicsData2_feat113","OmicsData2_feat139","OmicsData2_feat165","OmicsData2_feat185","OmicsData2_feat192"],"label":["OmicsData1_feat95","OmicsData2_feat105","OmicsData2_feat112","OmicsData2_feat113","OmicsData2_feat139","OmicsData2_feat165","OmicsData2_feat185","OmicsData2_feat192"],"color":["#3C0F71","#932B80","#932B80","#932B80","#932B80","#932B80","#932B80","#932B80"],"Dataset":["OmicsData1","OmicsData2","OmicsData2","OmicsData2","OmicsData2","OmicsData2","OmicsData2","OmicsData2"],"shadow":[true,true,true,true,true,true,true,true],"borderWidth":[1,1,1,1,1,1,1,1],"value":[7,1,1,1,1,1,1,1],"x":[-0.0357079167870175,-1,-0.331813144247507,-0.93159564760773,-0.1910311561671,1,0.665892952277266,0.554441218893547],"y":[0.0165915722297145,0.39883085474345,1,-0.493729165282298,-1,0.0891261697062371,-0.737517232511534,0.861809559979005]},"edges":{"from":["OmicsData2_feat192","OmicsData2_feat165","OmicsData2_feat113","OmicsData2_feat139","OmicsData2_feat185","OmicsData2_feat112","OmicsData2_feat105"],"to":["OmicsData1_feat95","OmicsData1_feat95","OmicsData1_feat95","OmicsData1_feat95","OmicsData1_feat95","OmicsData1_feat95","OmicsData1_feat95"],"width":[1.78344006976728,-1.47815198343782,-1.43848198410314,1.51542586667335,-1.10699685831388,-1.27453771707519,-1.03681736408471],"color":["red","blue","blue","red","blue","blue","blue"],"smooth":[false,false,false,false,false,false,false],"shadow":[false,false,false,false,false,false,false]},"nodesToDataframe":true,"edgesToDataframe":true,"options":{"width":"100%","height":"100%","nodes":{"shape":"dot","physics":false},"manipulation":{"enabled":false},"edges":{"smooth":false},"physics":{"stabilization":false}},"groups":null,"width":null,"height":null,"idselection":{"enabled":true,"style":"width: 150px; height: 26px","useLabels":true,"main":"Select by id"},"byselection":{"enabled":true,"style":"width: 150px; height: 26px","multiple":false,"hideColor":"rgba(200,200,200,0.5)","highlight":false,"variable":"Dataset","main":"Select by Dataset","values":["OmicsData1","OmicsData2"]},"main":null,"submain":null,"footer":null,"background":"rgba(0, 0, 0, 0)","highlight":{"enabled":true,"hoverNearest":false,"degree":1,"algorithm":"all","hideColor":"rgba(200,200,200,0.5)","labelOnly":true},"collapse":{"enabled":false,"fit":false,"resetHighlight":true,"clusterOptions":null,"keepCoord":true,"labelSuffix":"(cluster)"},"igraphlayout":{"type":"square"}},"evals":[],"jsHooks":[]}

# Set a lower threshold for the correlations to be shown as edges:
SPEARobj$plot.network(correlation.cutoff = .2)

## No argument passed for correlation.matrix, getting correlation matrix default (for Factor 1)...
## Getting correlation matrix for Factor 1
## Found 32 features for Factor 1
## Beginning pearson correlation tests for 32 features
## 0----------50--------100%
##  ====================

# Allow connections between the same dataset (omic)
SPEARobj$plot.network(allow.same.dataset = TRUE, 
                      correlation.cutoff = .4)

## No argument passed for correlation.matrix, getting correlation matrix default (for Factor 1)...
## Getting correlation matrix for Factor 1
## Found 32 features for Factor 1
## Beginning pearson correlation tests for 32 features
## 0----------50--------100%
##  ====================

{"x":{"nodes":{"id":["OmicsData3_feat120","OmicsData3_feat124","OmicsData3_feat21","OmicsData3_feat45","OmicsData3_feat57","OmicsData3_feat78","OmicsData3_feat89","OmicsData4_feat117","OmicsData4_feat170","OmicsData4_feat171","OmicsData4_feat194"],"label":["OmicsData3_feat120","OmicsData3_feat124","OmicsData3_feat21","OmicsData3_feat45","OmicsData3_feat57","OmicsData3_feat78","OmicsData3_feat89","OmicsData4_feat117","OmicsData4_feat170","OmicsData4_feat171","OmicsData4_feat194"],"color":["#DB5B68","#DB5B68","#DB5B68","#DB5B68","#DB5B68","#DB5B68","#DB5B68","#FAA138","#FAA138","#FAA138","#FAA138"],"Dataset":["OmicsData3","OmicsData3","OmicsData3","OmicsData3","OmicsData3","OmicsData3","OmicsData3","OmicsData4","OmicsData4","OmicsData4","OmicsData4"],"shadow":[true,true,true,true,true,true,true,true,true,true,true],"borderWidth":[1,1,1,1,1,1,1,1,1,1,1],"value":[2,1,3,2,2,2,1,1,1,1,8],"x":[-0.709187272203205,-1,-0.520411269475154,0.289764620503925,-0.494922565085773,-0.105824106405565,0.781309301630175,0.441162833775818,1,0.675113015423402,0.111180690583415],"y":[-0.00121017116217703,1,0.607381017363191,-0.540833533506206,-0.210647532365248,0.488436786383891,-0.175048825736814,-1,0.157305167490733,0.465971059159656,0.0893223396480152]},"edges":{"from":["OmicsData3_feat21","OmicsData3_feat57","OmicsData3_feat45","OmicsData3_feat78","OmicsData3_feat78","OmicsData4_feat170","OmicsData3_feat124","OmicsData3_feat89","OmicsData3_feat120","OmicsData3_feat120","OmicsData4_feat117","OmicsData4_feat171"],"to":["OmicsData4_feat194","OmicsData4_feat194","OmicsData4_feat194","OmicsData4_feat194","OmicsData3_feat21","OmicsData4_feat194","OmicsData3_feat21","OmicsData4_feat194","OmicsData4_feat194","OmicsData3_feat57","OmicsData3_feat45","OmicsData4_feat194"],"width":[-2.66150985311116,2.42701521898015,-2.37623365206968,-2.60176895564453,2.30657291758065,-2.11497853721,-2.08321723614541,2.1027020372372,-1.64356121334489,-1.39997893260334,1.52773274234317,1.43200921065679],"color":["blue","red","blue","blue","red","blue","blue","red","blue","blue","red","red"],"smooth":[false,false,false,false,false,false,false,false,false,false,false,false],"shadow":[false,false,false,false,false,false,false,false,false,false,false,false]},"nodesToDataframe":true,"edgesToDataframe":true,"options":{"width":"100%","height":"100%","nodes":{"shape":"dot","physics":false},"manipulation":{"enabled":false},"edges":{"smooth":false},"physics":{"stabilization":false}},"groups":null,"width":null,"height":null,"idselection":{"enabled":true,"style":"width: 150px; height: 26px","useLabels":true,"main":"Select by id"},"byselection":{"enabled":true,"style":"width: 150px; height: 26px","multiple":false,"hideColor":"rgba(200,200,200,0.5)","highlight":false,"variable":"Dataset","main":"Select by Dataset","values":["OmicsData3","OmicsData4"]},"main":null,"submain":null,"footer":null,"background":"rgba(0, 0, 0, 0)","highlight":{"enabled":true,"hoverNearest":false,"degree":1,"algorithm":"all","hideColor":"rgba(200,200,200,0.5)","labelOnly":true},"collapse":{"enabled":false,"fit":false,"resetHighlight":true,"clusterOptions":null,"keepCoord":true,"labelSuffix":"(cluster)"},"igraphlayout":{"type":"square"}},"evals":[],"jsHooks":[]}

`plot.coefficients`

Plot the coefficients of a SPEAR model.

Parameters:

coef.type Which coefficients to plot? Can be "projection" (default) or "regression"
plot.type Which type of plot? Defaults to "heatmap". (currently working on other plot.types)
factors Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL (all).
datasets Which datasets? Check names via names(SPEARobject$data$train$Xlist). Defaults to NULL (all)
cluster Should features be clustered? Defaults to TRUE
dendrogram.scale How large should the dendrogram from cluster be compared to each factor? Defaults to 1 (make > 1 for larger dendrogram).
show.all.names Should all feature names be shown? Defaults to FALSE (if num.features > 100, can be very crowded).
abbrev.names Should feature names be abbreviated to 20 characters? Defaults to TRUE
dark Use the dark theme? Defaults to FALSE

Examples:

SPEARobj$plot.coefficients()

SPEARobj$plot.coefficients(cluster = FALSE)

SPEARobj$plot.coefficients(cluster = FALSE, dark = TRUE)

SPEARobj$plot.coefficients(coef.type = "regression")

`plot.probabilities`

Plot the probabilities of a SPEAR model.

Parameters:

prob.type Which probabilities to plot? Can be "joint" (default), "nonzero", "marginal", or "projection"
plot.type Which type of plot? Defaults to "heatmap". (currently working on other plot.types)
factors Which factors to return features for? Accepts integers (i.e. factors = c(1,2,4) or factors = 1) Defaults to NULL (all).
datasets Which datasets? Check names via names(SPEARobject$data$train$Xlist). Defaults to NULL (all)
cluster Should features be clustered? Defaults to TRUE
dendrogram.scale How large should the dendrogram from cluster be compared to each factor? Defaults to 1 (make > 1 for larger dendrogram).
show.all.names Should all feature names be shown? Defaults to FALSE (if num.features > 100, can be very crowded).
abbrev.names Should feature names be abbreviated to 20 characters? Defaults to TRUE
log.probabilities Should the probabilites return the -log10(probabilities)? Defaults to TRUE
dark Use the dark theme? Defaults to FALSE

Examples:

SPEARobj$plot.probabilities()

SPEARobj$plot.probabilities(cluster = FALSE)

SPEARobj$plot.probabilities(log.probabilities = FALSE)

SPEARobj$plot.probabilities(prob.type = "nonzero")

Other Vignettes

To return to the main SPEAR vignette, click here

jgygi/SPEAR documentation built on July 5, 2023, 5:35 p.m.

rdrr.io home R language documentation Run R code online

CRAN packages Bioconductor packages R-Forge packages GitHub packages

Note that we can't provide technical support on individual packages. You should contact the package authors for that.

jgygi/SPEAR
SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

docs/vignettes/05_SPEAR_plots.md
In jgygi/SPEAR: SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

Plotting results from a SPEARobject

Required Libraries:

Overview:

`plot.factor.scores`

`plot.embedding`

`plot.contributions`

`plot.feature.expression`

`plot.features`

`plot.cv.loss`

`plot.factor.correlations`

`plot.predictions`

`plot.circos`

`plot.network`

`plot.coefficients`

`plot.probabilities`

Other Vignettes

R Package Documentation

Browse R Packages

We want your feedback!

jgygi/SPEAR SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

docs/vignettes/05_SPEAR_plots.md In jgygi/SPEAR: SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

Plotting results from a SPEARobject

Required Libraries:

Overview:

plot.factor.scores

plot.embedding

plot.contributions

plot.feature.expression

plot.features

plot.cv.loss

plot.factor.correlations

plot.predictions

plot.circos

plot.network

plot.coefficients

plot.probabilities

Other Vignettes

R Package Documentation

Browse R Packages

We want your feedback!

jgygi/SPEAR
SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

docs/vignettes/05_SPEAR_plots.md
In jgygi/SPEAR: SPEAR - Sparse Supervised Bayesian Factor Model for Multi-omic Analyses

`plot.factor.scores`

`plot.embedding`

`plot.contributions`

`plot.feature.expression`

`plot.features`

`plot.cv.loss`

`plot.factor.correlations`

`plot.predictions`

`plot.circos`

`plot.network`

`plot.coefficients`

`plot.probabilities`