The crisprShiny
package allows users to interact with
and visualize CRISPR guide RNAs (gRNAs) stored in GuideSet
objects via Shiny applications. This is possible either directly with a
self-contained app, or as a module within a larger app. This vignette
will demonstrate how to use this package to build Shiny applications for
GuideSet
objects, and how to navigate gRNA annotation
within the app.
Shiny apps created using the crisprShiny
package require
GuideSet
objects be constructed and annotated using the
crisprDesign
package, however, custom mcols
may be added to the GuideSet
, as shown below. For more
information about how to store and annotate CRISPR gRNAs with
GuideSet
objects, see our crisprVerse tutorials
page.
crisprShiny
can be installed from from the Bioconductor
devel branch using the following commands in a fresh R session:
The GuideSet
object can include a great variety of gRNA
annotation (see crisprDesign
package) that can be neatly
displayed in a Shiny app by simply passing it to the
crisprShiny
function. This function dynamically renders
components to display annotations that are present in the
GuideSet
. To demonstrate, let’s first use it with a
precomputed, non-annotated GuideSet
of gRNAs targeting the
CDS of the human KRAS gene, which is stored in the
crisprShiny
package. Note that the crisprShiny
function returns an app object, which can then be run using
shiny::runApp
.
if (interactive()){
library(crisprShiny)
data("guideSetExample_basic", package="crisprShiny")
app <- crisprShiny(guideSetExample_basic)
shiny::runApp(app)
}
Passing this GuideSet
returns a simple data table and a
button to set filtering criteria. The columns in the Table of
On-targets reflect the mcols present in the
GuideSet
(with an exception for certain list-column
annotations, such as alignments
, shown below). We can add
columns to the mcols
of the GuideSet
, which
then appear in the data table. (Note: adding columns that share names
with annotations added by functions in the crisprDesign
package can result in unexpected behaviors, and is therefore not
advised.)
if (interactive()){
gs <- guideSetExample_basic[1:50]
## add custom data columns
library <- c("lib1", "lib2", "lib3", "lib4")
set.seed(1000)
values <- round(rnorm(length(gs)), 4)
mcols(gs)[["values"]] <- values
library <- sample(library, length(gs), replace=TRUE)
mcols(gs)[["library"]] <- library
## create and run app
app <- crisprShiny(gs, geneModel=txdb_kras)
shiny::runApp(app)
}
For a more interesting app, we can pass a GuideSet
containing gRNA annotations to the crisprShiny
function.
This will include additional components for each in-depth gRNA-level
annotation, such as alignments, and gene annotation. For our example we
will use a fully-annotated version of the GuideSet
used
above.
if (interactive()){
data("guideSetExample_kras", package="crisprShiny")
app <- crisprShiny(guideSetExample_kras)
shiny::runApp(app)
}
Passing a fully-annotated GuideSet
to the
crisprShiny
function creates what can be thought of as a
“fully-functional” app, in that all annotation features are presented in
the app. Each are described in the following sections.
All mcols
in the GuideSet
that are not
list-columns (alignments
, etc.) are shown in this table,
with ID
, (names of the GuideSet
) given in the
first column. Some columns added by crisprDesign
functions
(see crisprDesign
package) have special formatting:
score_
and
supported by the crisprScore
package) have blue bars that
allow for quick comparison across gRNAs.n0
,
n1_c
, etc.) are blank for values of 0
, and
grayed out for NA
values (see
?crisprDesign::addSpacerAlignments
).TRUE
or FALSE
values, such as inRepeats
and polyT
) are
highlighted when the flag is present and blank otherwise, to quickly
identify potentially problematic gRNA features.There are several ways to interact with these data in the table:
Clicking on the
Filter
on-targets button opens the menu to filter gRNAs in the
on-targets table. Filters options depend on which annotations are
present in the GuideSet
object (see
crisprDesign
package for information on annotations), and
are divided into separate gRNA-feature categories, as described in the
following sections. The default filter values can be controlled with the
useFilterPresets
argument of the crisprShiny
function (see ?crisprShiny
for details).
This covers nucleotide-specific features, including
These filters set the upper limits for number of on- and off-targets
in the host genome. Limits can be set for each off-target by mismatch
count and whether the off-target occurs in a gene CDS or promoter
region. The GuideSet
must have alignment annotation for
this section to be available.
This section sets the lower limits for on- and off-target scores, as
well as options for permitting NA
scores (such as with
non-targeting control sequences). See the crisprScores
package for available scoring methods.
This section concerns the genomic context of protospacers: whether they overlap repeat elements, SNPs, or Pfam domains.
This section allows for narrowing gRNAs to those targeting specific isoforms, a defined region within that isoform, and a minimum percent of isoform coverage for the target gene.
This section filters for gRNAs that target, and are within a certain distance from, a specific TSS.
As stated in component, the user may select up to 20 gRNAs from same
chromosome in the table of on-targets to plot using the
crisprViz
package. In this example, no value was passed to
the geneModel
argument, so no gene information is available
for the plot. If we rerun the function with a GRangesList
(see crisprDesign
package), such as txdb_kras
,
the gene transcripts are also plotted. There are additional options to
set the viewing range and how the gRNA track is organized.
Regardless of what value is passed to geneModel
, this
component is only available if the GuideSet
is annotated
with gene_symbol
and/or gene_id
column(s) in
mcols
.
Additonal interactive components for detailed annotations present in
the GuideSet
can be found at the bottom of the app.
Annotations are presented one gRNA at a time, selected using the
dropdown menu, and are described below.
This tab presents information about each on- and off-target. The putative protospacers can be filtered by maximum number of mismatches, target genomic context, and whether they are adjacent to canonical PAMs, as well as sorted by CFD or MIT score (SpCas9 only).
Just below the filters is a quick summary of the on- and off-target count for the gRNA, and details about each putative protospacer in a datatable, such as where mismatches occur, off-target scores, and its genomic location with gene-level context. Rows in the datatable can be selected to visualize the protospacer target with respect to the target or nearby gene (up to 10kb).
Currently, this tab displays a datatable of gene annotation for the
selected gRNA, identical to the output of
crisprShiny::geneAnnotation(guideSet)
.
Currently, this tab displays a datatable of TSS annotation for the
selected gRNA, identical to the output of
crisprShiny::tssAnnotation(guideSet)
.
This tab includes a datatable of restriction enzymes whose motifs are
found in the gRNA. Restriction enzymes in the table and gRNA flanking
sequences used to determine presence of motifs depend on arguments
passed to crisprDesign::addRestrictionEnzymes
in
constructing the input GuideSet
.
Currently, this tab displays a datatable of SNPs annotation for the
selected gRNA, identical to the output of
crisprShiny::snps(guideSet)
.
The UI and server components of crisprShiny
are also
available outside of the crisprShiny
function to be used as
modules within a larger app that, for instance, may allow the user to
parse multiple GuideSet
objects in a single session. Below
is a simple example that allows the user to explore both the basic and
fully-annotated GuideSet
s used above in the same
session.
if (interactive()){
library(shiny)
ui <- function(id){
fluidPage(
sidebarLayout(
sidebarPanel(
selectInput(
inputId="guideSet_select",
label="Select GuideSet",
choices=c("simple", "full")
)
),
mainPanel(
crisprUI(id)
)
)
)
}
server <- function(id){
function(input, output, session){
guideSet_list <- list("simple"=guideSetExample_basic,
"full"=guideSetExample_kras[1:50])
gs <- reactive({
req(input$guideSet_select)
guideSet_list[[input$guideSet_select]]
})
observeEvent(gs(), {
crisprServer(
id,
guideSet=gs(),
geneModel=txdb_kras,
useFilterPresets=TRUE
)
})
}
}
myApp <- function(){
id <- "id"
shinyApp(ui=ui(id), server=server(id))
}
myApp()
}
## R version 4.4.2 (2024-10-31)
## Platform: x86_64-pc-linux-gnu
## Running under: Ubuntu 24.04.1 LTS
##
## Matrix products: default
## BLAS: /usr/lib/x86_64-linux-gnu/openblas-pthread/libblas.so.3
## LAPACK: /usr/lib/x86_64-linux-gnu/openblas-pthread/libopenblasp-r0.3.26.so; LAPACK version 3.12.0
##
## locale:
## [1] LC_CTYPE=en_US.UTF-8 LC_NUMERIC=C
## [3] LC_TIME=en_US.UTF-8 LC_COLLATE=C
## [5] LC_MONETARY=en_US.UTF-8 LC_MESSAGES=en_US.UTF-8
## [7] LC_PAPER=en_US.UTF-8 LC_NAME=C
## [9] LC_ADDRESS=C LC_TELEPHONE=C
## [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C
##
## time zone: Etc/UTC
## tzcode source: system (glibc)
##
## attached base packages:
## [1] stats graphics grDevices utils datasets methods base
##
## other attached packages:
## [1] BiocStyle_2.35.0
##
## loaded via a namespace (and not attached):
## [1] cli_3.6.3 knitr_1.49 rlang_1.1.4
## [4] xfun_0.49 promises_1.3.2 shiny_1.10.0
## [7] xtable_1.8-4 jsonlite_1.8.9 buildtools_1.0.0
## [10] htmltools_0.5.8.1 maketools_1.3.1 httpuv_1.6.15
## [13] sys_3.4.3 sass_0.4.9 rmarkdown_2.29
## [16] evaluate_1.0.1 jquerylib_0.1.4 fontawesome_0.5.3
## [19] fastmap_1.2.0 yaml_2.3.10 lifecycle_1.0.4
## [22] BiocManager_1.30.25 compiler_4.4.2 Rcpp_1.0.13-1
## [25] later_1.4.1 digest_0.6.37 R6_2.5.1
## [28] magrittr_2.0.3 bslib_0.8.0 tools_4.4.2
## [31] mime_0.12 cachem_1.1.0