Bioconductor

Project

Release News

• 3.9 release
• 3.8 release
• 3.7 release
• 3.6 release
• 3.5 release

Annual reports

http://bioconductor.org/about/annual-reports/

Download stats

• See the overview vignette of BiocPkgTools
• bioconductor.riken.jp mirror in Japan
  • Download stats for Bioconductor software packages
  • Download stats for Bioconductor annotation packages
• biocpkg package in github

From the director of the project

useR 2019 & youtube

Publications

https://www.bioconductor.org/help/publications/

Bioconductor User Support

https://support.bioconductor.org/

Mirrors

https://www.bioconductor.org/about/mirrors/

Github mirror

• https://support.bioconductor.org/p/68824/ Announcement (Update: it is dead)

Japan

http://bioconductor.jp

Package source

• R > create Bioconductor repository

Code search

http://search.bioconductor.jp/

Resource

Teaching resources from rafalab

BiocManager from CRAN

The reason for using BiocManager instead of biocLite() is mostly to stop sourcing an R script from URL which isn’t so safe. So biocLite() should not be recommended anymore.

It allows to have multiple versions of Bioconductor installed on the same computer. For example, R 3.5 works with Bioconductor 3.7 and 3.8.

On the other hand, setRepositories(ind=1:4) and install.packages() still lets you install Bioconductor packages.

Hacking Bioconductor

BiocPkgTools

• BiocPkgTools: Collection of simple tools for learning about Bioc Packages
• https://www.biorxiv.org/content/10.1101/642132v1
• GitHub Action executed as a CRON job to update the website daily. Note that the 'bioc' version needs to be updated once a new version of Bioconductor has been released.

BioCExplorer

Explore Bioconductor packages more nicely

source("https://bioconductor.org/biocLite.R")
biocLite("BiocUpgrade")
biocLite("biocViews")
devtools::install_github("seandavi/BiocPkgTools")
devtools::install_github("shians/BioCExplorer")
library(BioCExplorer)
bioc_explore()

BiocViews

• Software
  • AssayDomain
  • BiologicalQuestion
  • Infrastructure
  • ResearchField
  • StatisticalMethod
  • Technology
  • WorkflowStep
• AnnotationData
  • ChipManufacturer
  • ChipName
  • CustomArray
  • CustomCDF
  • CustomDBSchema
  • FunctionalAnnotation
  • Organism
  • PackageType
  • SequenceAnnotation
• ExperimentData
  • AssayDomainData
  • DiseaseModel
  • OrganismData
  • PackageTypeData
  • RepositoryData
  • ReproducibleResearch
  • SpecimenSource
  • TechnologyData
• Workflow
  • AnnotationWorkflow
  • BasicWorkflow
  • DifferentialSplicingWorkflow
  • EpigeneticsWorkflow
  • GeneExpressionWorkflow
  • GenomicVariantsWorkflow
  • ImmunoOncologyWorkflow
  • ProteomicsWorkflow
  • ResourceQueryingWorkflow
  • SingleCellWorkflow

Annotation packages

• http://bioconductor.org/help/course-materials/2012/SeattleFeb2012/Annotation.pdf
• https://bioconductor.org/help/course-materials/2017/CSAMA/lectures/1-monday/lecture-04-a-annotation-intro/lecture-04a-annotation-intro.html
• Making and Utilizing TxDb Objects
• Genomic Annotation Resources Introduction to using gene, pathway, gene ontology, homology annotations and the AnnotationHub. Access GO, KEGG, NCBI, Biomart, UCSC, vendor, and other sources.
  • AnnotationHub allows us to query and download many different annotation objects without having to explicitly install them.
  • OrgDb
  • TxDb
  • OrganismDb packages are meta-packages that contain an OrgDb, a TxDb, and a GO.db packages and allow cross-queries between those packages.
  • BSgenome packages contain sequence information for a given species/build.
  • biomaRt allows queries to an Ensembl Biomart server.
• http://genomicsclass.github.io/book/pages/bioc1_annoCheat.html
• ensembldb package

• Introduction to Bioconductor annotation resources 2020 Bioc workshop
  • Functions

    select(annopkg, keys, columns, keytype): keytype is optional
    mapIds(annopkg, keys, column, keytype, multiVals): keytype is required, multiVals is used to handle duplicates.
    

  • Packages

    TxDb packages - eg TxDb.Hsapiens.UCSC.hg19.knownGene. It contains positional information.
    EnsDb packages - eg EnsDb.Hsapiens.v79
    

Gene centric

• AnnotationDbi: Introduction To Bioconductor Annotation Packages

library(hgu133a.db)
library(AnnotationDbi)
k <- head(keys(hgu133a.db, keytype="PROBEID"))
k
# [1] "1007_s_at" "1053_at"   "117_at"    "121_at"    "1255_g_at" "1294_at"
# then call select
select(hgu133a.db, keys=k, columns=c("SYMBOL","GENENAME"), keytype="PROBEID")

# 'select()' returned 1:many mapping between keys and columns
#     PROBEID  SYMBOL                                     GENENAME
# 1 1007_s_at    DDR1  discoidin domain receptor tyrosine kinase 1
# 2 1007_s_at MIR4640                                microRNA 4640
# 3   1053_at    RFC2               replication factor C subunit 2
# 4    117_at   HSPA6 heat shock protein family A (Hsp70) member 6
# 5    121_at    PAX8                                 paired box 8
# 6 1255_g_at  GUCA1A               guanylate cyclase activator 1A
# 7   1294_at    UBA7  ubiquitin like modifier activating enzyme 7
# 8   1294_at MIR5193                                microRNA 5193

Genomic centric

• GRanges()
• GrangesList()

GRanges() and GRangesLists act like data.frames and lists, and can be subsetted using the [ function.

SummarizedExperiment objects

SummarizedExperiment objects are like ExpressionSets, but the row-wise annotations are GRanges, so you can subset by genomic locations.

SummarizedExperiment objects are popular objects for representing expression data and other rectangular data (feature x sample data).

Some packages

Biobase, GEOquery and limma

How to create an ExpressionSet object from scratch? Here we use the code from GEO2R to help to do this task.

library(Biobase)
library(GEOquery)
library(limma)

# Load series and platform data from GEO

gset <- getGEO("GSE32474", GSEMatrix =TRUE, AnnotGPL=TRUE)
if (length(gset) > 1) idx <- grep("GPL570", attr(gset, "names")) else idx <- 1
gset <- gset[[idx]]
# save(gset, file = "~/Downloads/gse32474_gset.rda")
# load("~/Downloads/gse32474_gset.rda")
table(pData(gset)[, "cell line:ch1"])
pData(gset)

# Create an ExpressionSet object from scratch
# We take a shortcut to obtain the pheno data and feature data matrices 
# from the output of getGEO()
phenoDat <- new("AnnotatedDataFrame",
                 data=pData(gset))
featureDat <- new("AnnotatedDataFrame",
                  data=fData(gset))
exampleSet <- ExpressionSet(assayData=exprs(gset),
                            phenoData=phenoDat,
                            featureData=featureDat,
                            annotation="hgu133plus2")
gset <- exampleSet

# Make proper column names to match toptable 
fvarLabels(gset) <- make.names(fvarLabels(gset))

# group names for all samples
gsms <- paste0("00000000111111111XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
               "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
               "XXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXXX",
               "XXXXXXXXXXXXXXXXXXXXXXXX")
sml <- c()
for (i in 1:nchar(gsms)) { sml[i] <- substr(gsms,i,i) }

# Subset an ExpressionSet by eliminating samples marked as "X"
sel <- which(sml != "X")
sml <- sml[sel]
gset <- gset[ ,sel]

# Decide if it is necessary to do a log2 transformation
ex <- exprs(gset)
qx <- as.numeric(quantile(ex, c(0., 0.25, 0.5, 0.75, 0.99, 1.0), na.rm=T))
LogC <- (qx[5] > 100) ||
  (qx[6]-qx[1] > 50 && qx[2] > 0) ||
  (qx[2] > 0 && qx[2] < 1 && qx[4] > 1 && qx[4] < 2)
if (LogC) { ex[which(ex <= 0)] <- NaN
exprs(gset) <- log2(ex) }

# Set up the data and proceed with analysis
sml <- paste("G", sml, sep="")    # set group names
fl <- as.factor(sml)
gset$description <- fl
design <- model.matrix(~ description + 0, gset)
colnames(design) <- levels(fl)
fit <- lmFit(gset, design)
cont.matrix <- makeContrasts(G1-G0, levels=design)
fit2 <- contrasts.fit(fit, cont.matrix)
fit2 <- eBayes(fit2, 0.01)
tT <- topTable(fit2, adjust="fdr", sort.by="B", number=250)

# Display the result with selected columns
tT <- subset(tT, select=c("ID","adj.P.Val","P.Value","t","B","logFC","Gene.symbol","Gene.title"))
tT[1:2, ]
#                  ID  adj.P.Val      P.Value        t        B    logFC Gene.symbol
# 209108_at 209108_at 0.08400054 4.438757e-06 6.686977 3.786222 3.949088      TSPAN6
# 204975_at 204975_at 0.08400054 6.036355e-06 6.520775 3.550036 2.919995        EMP2
#                             Gene.title
# 209108_at                 tetraspanin 6
# 204975_at epithelial membrane protein 2

Biostrings

• Find the location of a particular sequence. ?vmatchPattern
• https://www.bioconductor.org/help/course-materials/2011/BioC2011/LabStuff/BiostringsBSgenomeOverview.pdf

library(Biostrings) 
library(BSgenome.Hsapiens.UCSC.hg19) 
vmatchPattern("GCGATCGC", Hsapiens)

plyranges

http://bioconductor.org/packages/devel/bioc/vignettes/plyranges/inst/doc/an-introduction.html

Misc

Package release history

https://support.bioconductor.org/p/69657/

Search the DESCRIPTION file (eg. VariantAnnotation package) in github and the release information can be found there.

Papers/Overview

Using R and Bioconductor in Clinical Genomics and Transcriptomics 2019