diff --git a/adam-r/bdgenomics.adam/DESCRIPTION b/adam-r/bdgenomics.adam/DESCRIPTION
index 18fb21c974..1ffc7d7e44 100644
--- a/adam-r/bdgenomics.adam/DESCRIPTION
+++ b/adam-r/bdgenomics.adam/DESCRIPTION
@@ -5,17 +5,17 @@ Title: R Frontend for Big Data Genomics/ADAM
 Description: ADAM is a genomics analysis platform with specialized file formats built using Apache Avro, Apache Spark and Parquet.
 Author: Big Data Genomics
 Maintainer: Frank Austin Nothaft <fnothaft@alumni.stanford.edu>
-Authors@R: c(person("Frank", "Nothaft", role = c("aut", "cre"),
+Authors@R: c(person("Frank Austin", "Nothaft", role = c("aut", "cre"),
                     email = "fnothaft@alumni.stanford.edu"),
              person(family = "Big Data Genomics", role = c("aut", "cph")))
 License: Apache License (== 2.0)
 URL: http://www.bdgenomics.org https://github.com/bigdatagenomics/adam
 BugReports: https://github.com/bigdatagenomics/adam/issues
 Imports:
-    methods
-Depends:
-    R (>= 3.0),
+    methods,
     SparkR (>= 2.1.0)
+Depends:
+    R (>= 3.0)
 Suggests:
     testthat
 Collate:
diff --git a/adam-r/bdgenomics.adam/NAMESPACE b/adam-r/bdgenomics.adam/NAMESPACE
index c496fce307..dc24dd4375 100644
--- a/adam-r/bdgenomics.adam/NAMESPACE
+++ b/adam-r/bdgenomics.adam/NAMESPACE
@@ -1,6 +1,40 @@
 # Generated by roxygen2: do not edit by hand
 
 export(ADAMContext)
+export(aggregatedCoverage)
+export(collapse)
+export(countKmers)
+export(coverage)
+export(createADAMContext)
+export(flankAdjacentFragments)
+export(flatten)
+export(loadAlignments)
+export(loadContigFragments)
+export(loadCoverage)
+export(loadFeatures)
+export(loadFragments)
+export(loadGenotypes)
+export(loadVariants)
+export(markDuplicates)
+export(pipe)
+export(realignIndels)
+export(recalibrateBaseQualities)
+export(save)
+export(saveAsParquet)
+export(saveAsSam)
+export(saveAsVcf)
+export(sort)
+export(sortLexicographically)
+export(sortReadsByReferencePosition)
+export(sortReadsByReferencePositionAndIndex)
+export(toCoverage)
+export(toDF)
+export(toFeatures)
+export(toFragments)
+export(toReads)
+export(toVariantContexts)
+export(transform)
+export(transmute)
 exportClasses(ADAMContext)
 exportClasses(AlignmentRecordRDD)
 exportClasses(CoverageRDD)
@@ -45,3 +79,8 @@ exportMethods(toReads)
 exportMethods(toVariantContexts)
 exportMethods(transform)
 exportMethods(transmute)
+importFrom(SparkR,sparkR.callJMethod)
+importFrom(SparkR,sparkR.callJStatic)
+importFrom(SparkR,sparkR.newJObject)
+importFrom(SparkR,sparkR.session)
+importFrom(methods,new)
diff --git a/adam-r/bdgenomics.adam/R/adam-context.R b/adam-r/bdgenomics.adam/R/adam-context.R
index 837ee8bbc4..4cf59ef93f 100644
--- a/adam-r/bdgenomics.adam/R/adam-context.R
+++ b/adam-r/bdgenomics.adam/R/adam-context.R
@@ -23,10 +23,32 @@ setOldClass("jobj")
 #' @description The ADAMContext provides helper methods for loading in genomic
 #'              data into a Spark RDD/Dataframe.
 #' @slot jac Java object reference to the backing JavaADAMContext.
+#'
+#' @rdname ADAMContext
+#' 
 #' @export
 setClass("ADAMContext",
          slots = list(jac = "jobj"))
 
+#' Creates an ADAMContext by creating a SparkSession.
+#'
+#' @return Returns an ADAMContext.
+#' 
+#' @importFrom SparkR sparkR.session
+#'
+#' @export
+createADAMContext <- function() {
+    ADAMContext(sparkR.session())
+}
+
+#' Creates an ADAMContext from an existing SparkSession.
+#'
+#' @param ss The Spark Session to use to create the ADAMContext.
+#' @return Returns an ADAMContext.
+#'
+#' @importFrom SparkR sparkR.callJMethod sparkR.newJObject
+#' @importFrom methods new
+#'
 #' @export
 ADAMContext <- function(ss) {
     ssc = sparkR.callJMethod(ss, "sparkContext")
@@ -36,6 +58,7 @@ ADAMContext <- function(ss) {
     new("ADAMContext", jac = jac)
 }
 
+#' @importFrom SparkR sparkR.callJStatic
 javaStringency <- function(stringency) {
     stringency <- sparkR.callJStatic("htsjdk.samtools.ValidationStringency",
                                      "valueOf",
@@ -62,6 +85,8 @@ javaStringency <- function(stringency) {
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing reads.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadAlignments",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -85,6 +110,8 @@ setMethod("loadAlignments",
 #' @param filePath The path to load the file from.
 #' @return Returns an RDD containing sequence fragments.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadContigFragments",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -109,6 +136,8 @@ setMethod("loadContigFragments",
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing sequence fragments.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadFragments",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -141,6 +170,8 @@ setMethod("loadFragments",
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing features.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadFeatures",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -174,6 +205,8 @@ setMethod("loadFeatures",
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing coverage.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadCoverage",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -196,6 +229,8 @@ setMethod("loadCoverage",
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing genotypes.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadGenotypes",
           signature(ac = "ADAMContext", filePath = "character"),
@@ -218,6 +253,8 @@ setMethod("loadGenotypes",
 #' @param stringency The validation stringency to apply. Defaults to STRICT.
 #' @return Returns an RDD containing variants.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("loadVariants",
           signature(ac = "ADAMContext", filePath = "character"),
diff --git a/adam-r/bdgenomics.adam/R/generics.R b/adam-r/bdgenomics.adam/R/generics.R
index b1f8b8ed0d..6a005567da 100644
--- a/adam-r/bdgenomics.adam/R/generics.R
+++ b/adam-r/bdgenomics.adam/R/generics.R
@@ -18,50 +18,71 @@
 
 #### ADAM Context operations ####
 
-# @rdname ADAMContext
-# @export
+#' The ADAMContext provides functions on top of a SparkContext for loading genomic data.
+#' 
+#' @name ADAMContext
+NULL
+
+#' @rdname ADAMContext
+#' @param ac The ADAMContext.
+#' @param filePath The path to load the file from.
+#' @param ... additional argument(s).
+#' @export
 setGeneric("loadAlignments",
            function(ac, filePath, ...) { standardGeneric("loadAlignments") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadContigFragments",
            function(ac, filePath) { standardGeneric("loadContigFragments") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadFragments",
            function(ac, filePath, ...) { standardGeneric("loadFragments") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadFeatures",
            function(ac, filePath, ...) { standardGeneric("loadFeatures") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadCoverage",
            function(ac, filePath, ...) { standardGeneric("loadCoverage") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadGenotypes",
            function(ac, filePath, ...) { standardGeneric("loadGenotypes") })
 
-# @rdname ADAMContext
-# @export
+#' @rdname ADAMContext
+#' @export
 setGeneric("loadVariants",
            function(ac, filePath, ...) { standardGeneric("loadVariants") })
 
 #### RDD operations ####
 
-# @rdname GenomicRDD
-# @export
+#' The GenomicRDD is the base class that all genomic datatypes extend from in ADAM.
+#' 
+#' @name GenomicRDD
+NULL
+
+#' @rdname GenomicRDD
+#' @param cmd The command to run.
+#' @param tFormatter The name of the ADAM in-formatter class to use.
+#' @param xFormatter The name of the ADAM out-formatter class to use.
+#' @param convFn The name of the ADAM GenomicRDD conversion class to
+#'   use.
+#' @param ... additional argument(s).
+#' @return Returns a new RDD where the input from the original RDD has
+#'   been piped through a command that runs locally on each executor.
+#' @export
 setGeneric("pipe",
            function(ardd, cmd, tFormatter, xFormatter, convFn, ...) { standardGeneric("pipe") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @export
 setGeneric("toDF",
            function(ardd) { standardGeneric("toDF") })
 
@@ -71,8 +92,8 @@ setGeneric("replaceRdd",
 setGeneric("wrapTransformation",
            function(ardd, tFn) { standardGeneric("wrapTransformation") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @export
 setGeneric("transform",
            function(ardd, tFn) { standardGeneric("transform") })
 
@@ -82,130 +103,167 @@ setGeneric("inferConversionFn",
 setGeneric("destClassSuffix",
            function(destClass) { standardGeneric("destClassSuffix") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @param tFn A function that transforms the underlying RDD as a DataFrame.
+#' @param destClass The destination class of this transmutation.
+#' @export
 setGeneric("transmute",
            function(ardd, tFn, destClass, ...) { standardGeneric("transmute") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @export
 setGeneric("save",
            function(ardd, filePath, ...) { standardGeneric("save") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @export
 setGeneric("sort",
            function(ardd) { standardGeneric("sort") })
 
-# @rdname GenomicRDD
-# @export
+#' @rdname GenomicRDD
+#' @export
 setGeneric("sortLexicographically",
            function(ardd) { standardGeneric("sortLexicographically") })
 
+#' Saves this RDD to disk as Parquet.
+#'
+#' @param ardd The RDD to apply this to.
+#' @param filePath Path to save file to.
+#'
+#' @rdname GenomicRDD
+#'
+#' @export
+setGeneric("saveAsParquet",
+           function(ardd, filePath) { standardGeneric("saveAsParquet") })
+
 #### AlignmentRecord operations ####
 
-# @rdname AlignmentRecordRDD
-# @export
+#' The AlignmentRecordRDD is the class used to manipulate genomic read data.
+#' 
+#' @name AlignmentRecordRDD
+NULL
+
+#' @rdname AlignmentRecordRDD
+#' @export
 setGeneric("toFragments",
            function(ardd) { standardGeneric("toFragments") })
 
-# @rdname AlignmentRecordRDD
-# @export
+#' @rdname AlignmentRecordRDD
+#' @param ardd The RDD to apply this to.
+#' @param ... additional argument(s).
+#' @export
 setGeneric("toCoverage",
            function(ardd, ...) { standardGeneric("toCoverage") })
 
-# @rdname AlignmentRecordRDD
-# @export
+#' @rdname AlignmentRecordRDD
+#' @param kmerLength The value of _k_ to use for cutting _k_-mers.
+#' @export
 setGeneric("countKmers",
            function(ardd, kmerLength) { standardGeneric("countKmers") })
 
-# @rdname AlignmentRecordRDD
-# @export
+#' @rdname AlignmentRecordRDD
+#' @param filePath The path to save the file to.
+#' @export
 setGeneric("saveAsSam",
            function(ardd, filePath, ...) { standardGeneric("saveAsSam") })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
+#' @rdname AlignmentRecordRDD
+#' @export
 setGeneric("sortReadsByReferencePosition",
            function(ardd) { standardGeneric("sortReadsByReferencePosition") })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
+#' @rdname AlignmentRecordRDD
+#' @export
 setGeneric("sortReadsByReferencePositionAndIndex",
            function(ardd) { standardGeneric("sortReadsByReferencePositionAndIndex") })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
+#' @rdname AlignmentRecordRDD
+#' @export
 setGeneric("markDuplicates",
            function(ardd) { standardGeneric("markDuplicates") })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
+#' @rdname AlignmentRecordRDD
+#' @param knownSnps A table of known SNPs to mask valid variants.
+#' @param validationStringency The stringency to apply towards validating BQSR.
+#' @export
 setGeneric("recalibrateBaseQualities",
            function(ardd, knownSnps, validationStringency) {
              standardGeneric("recalibrateBaseQualities")
            })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
+#' @rdname AlignmentRecordRDD
+#' @export
 setGeneric("realignIndels",
            function(ardd, ...) { standardGeneric("realignIndels") })
 
-# @rdname AlignmentRecordRDD-transforms
-# @export
-setGeneric("realignIndels",
-           function(ardd, knownIndels, ...) { standardGeneric("realignIndels") })
-
 #### Coverage operations ####
 
-# @rdname CoverageRDD
-# @export
+#' The CoverageRDD class is used to manipulate read coverage counts.
+#' 
+#' @name CoverageRDD
+NULL
+
+#' @rdname CoverageRDD
+#' @param ... additional argument(s).
+#' @export
 setGeneric("collapse",
            function(ardd, ...) { standardGeneric("collapse") })
 
-# @rdname CoverageRDD
-# @export
+#' @rdname CoverageRDD
+#' @export
 setGeneric("toFeatures",
            function(ardd) { standardGeneric("toFeatures") })
 
-# @rdname CoverageRDD
-# @export
+#' @rdname CoverageRDD
+#' @export
 setGeneric("coverage",
            function(ardd, ...) { standardGeneric("coverage") })
 
-# @rdname CoverageRDD
-# @export
+#' @rdname CoverageRDD
+#' @export
+#' @aliases aggregatedCoverage,CoverageRDD-method
 setGeneric("aggregatedCoverage",
            function(ardd, ...) { standardGeneric("aggregatedCoverage") })
 
-# @rdname CoverageRDD
-# @export
+#' @rdname CoverageRDD
+#' @export
 setGeneric("flatten",
            function(ardd) { standardGeneric("flatten") })
 
 #### Fragment operations ####
 
-# @rdname FragmentRDD
-# @export
+#' The FragmentRDD class is used to manipulate paired reads.
+#' 
+#' @name FragmentRDD
+NULL
+
+#' @rdname FragmentRDD
+#' @param ardd The RDD to apply this to.
+#' @export
 setGeneric("toReads",
            function(ardd) { standardGeneric("toReads") })
 
-#### Genotype operations ####
+#### Genotype and Variant operations ####
 
-# @rdname GenotypeRDD
-# @export
+#' Converts this RDD to VariantContexts.
+#'
+#' @param ardd The RDD to apply this to.
+#' @return Returns this RDD of Variants as VariantContexts.
+#' @export
 setGeneric("toVariantContexts",
            function(ardd) { standardGeneric("toVariantContexts") })
 
-# @rdname GenotypeRDD
-# @export
-setGeneric("saveAsParquet",
-           function(ardd, filePath) { standardGeneric("saveAsParquet") })
-
 #### NucleotideContigFragment operations ####
 
-# @rdname NucleotideContigFragmentRDD
-# @export
+#' The NucleotideContigFragmentRDD class is used to manipulate contigs.
+#' 
+#' @name NucleotideContigFragmentRDD
+NULL
+
+#' @rdname NucleotideContigFragmentRDD
+#' @param ardd The RDD to apply this to.
+#' @param flankLength The length to extend adjacent records by.
+#' @export
 setGeneric("flankAdjacentFragments",
            function(ardd, flankLength) {
              standardGeneric("flankAdjacentFragments")
@@ -213,17 +271,20 @@ setGeneric("flankAdjacentFragments",
 
 #### Variant operations ####
 
-# @rdname VariantRDD
-# @export
-setGeneric("toVariantContexts",
-           function(ardd) { standardGeneric("toVariantContexts") })
-
-# @rdname VariantRDD
-# @export
-setGeneric("saveAsParquet",
-           function(ardd, filePath) { standardGeneric("saveAsParquet") })
-
-# @rdname VariantContextRDD
-# @export
+#' The VariantContextRDD class is used to manipulate VCF-styled data.
+#'
+#' Each element in a VariantContext RDD corresponds to a VCF line. This
+#' differs from the GenotypeRDD, where each element represents the genotype
+#' of a single sample at a single site, or a VariantRDD, which represents
+#' just the variant of interest.
+#' 
+#' @name VariantContextRDD
+NULL
+
+#' @rdname VariantContextRDD
+#' @param ardd The RDD to apply this to.
+#' @param filePath Path to save VCF to.
+#' @param ... additional argument(s).
+#' @export
 setGeneric("saveAsVcf",
            function(ardd, filePath, ...) { standardGeneric("saveAsVcf") })
diff --git a/adam-r/bdgenomics.adam/R/rdd.R b/adam-r/bdgenomics.adam/R/rdd.R
index 56269bc4df..83b09af51e 100644
--- a/adam-r/bdgenomics.adam/R/rdd.R
+++ b/adam-r/bdgenomics.adam/R/rdd.R
@@ -19,44 +19,75 @@ library(SparkR)
 
 setOldClass("jobj")
 
+#' A class that wraps an RDD of genomic data with helpful metadata.
+#'
+#' @rdname GenomicRDD
+#' @slot jrdd The Java RDD that this class wraps.
+#' 
 #' @export
 setClass("GenomicRDD",
          slots = list(jrdd = "jobj"))
 
-
+#' A class that wraps a DataFrame of genomic data with helpful metadata.
+#'
+#' @rdname GenomicDataset
+#' @slot jrdd The Java RDD that this class wraps.
+#' 
 #' @export
 setClass("GenomicDataset",
          slots = list(jrdd = "jobj"),
          contains = "GenomicRDD")
 
-
+#' A class that wraps an RDD of genomic reads with helpful metadata.
+#'
+#' @rdname AlignmentRecordRDD
+#' @slot jrdd The Java RDD of AlignmentRecords that this class wraps.
+#' 
 #' @export
 setClass("AlignmentRecordRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 AlignmentRecordRDD <- function(jrdd) {
     new("AlignmentRecordRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of genomic coverage data with helpful metadata.
+#'
+#' @rdname CoverageRDD
+#' @slot jrdd The Java RDD of Coverage that this class wraps.
+#' 
 #' @export
 setClass("CoverageRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 CoverageRDD <- function(jrdd) {
     new("CoverageRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of genomic features with helpful metadata.
+#'
+#' @rdname FeatureRDD
+#' @slot jrdd The Java RDD of Features that this class wraps.
+#' 
 #' @export
 setClass("FeatureRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 FeatureRDD <- function(jrdd) {
     new("FeatureRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of read pairs grouped by sequencing fragment with helpful metadata.
+#'
+#' @rdname FragmentRDD
+#' @slot jrdd The Java RDD of Fragments that this class wraps.
+#' 
 #' @export
 setClass("FragmentRDD",
          slots = list(jrdd = "jobj"),
@@ -66,38 +97,62 @@ FragmentRDD <- function(jrdd) {
     new("FragmentRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of genotypes with helpful metadata.
+#'
+#' @rdname GenotypeRDD
+#' @slot jrdd The Java RDD of Genotypes that this class wraps.
+#' 
 #' @export
 setClass("GenotypeRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 GenotypeRDD <- function(jrdd) {
     new("GenotypeRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of contigs with helpful metadata.
+#'
+#' @rdname NucleotideContigFragmentRDD
+#' @slot jrdd The Java RDD of contigs that this class wraps.
+#' 
 #' @export
 setClass("NucleotideContigFragmentRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 NucleotideContigFragmentRDD <- function(jrdd) {
     new("NucleotideContigFragmentRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of variants with helpful metadata.
+#'
+#' @rdname VariantRDD
+#' @slot jrdd The Java RDD of Variants that this class wraps.
+#' 
 #' @export
 setClass("VariantRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicDataset")
 
+#' @importFrom methods new
 VariantRDD <- function(jrdd) {
     new("VariantRDD", jrdd = jrdd)
 }
 
+#' A class that wraps an RDD of both variants and genotypes with helpful metadata.
+#'
+#' @rdname VariantContextRDD
+#' @slot jrdd The Java RDD of VariantContexts that this class wraps.
+#' 
 #' @export
 setClass("VariantContextRDD",
          slots = list(jrdd = "jobj"),
          contains = "GenomicRDD")
 
+#' @importFrom methods new
 VariantContextRDD <- function(jrdd) {
     new("VariantContextRDD", jrdd = jrdd)
 }
@@ -116,6 +171,7 @@ VariantContextRDD <- function(jrdd) {
 #' format the input to the pipe, and the implicit OutFormatter is used to
 #' parse the output from the pipe.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param cmd The command to run.
 #' @param tFormatter The name of the ADAM in-formatter class to use.
 #' @param xFormatter The name of the ADAM out-formatter class to use.
@@ -130,6 +186,8 @@ VariantContextRDD <- function(jrdd) {
 #' @return Returns a new RDD where the input from the original RDD has
 #'   been piped through a command that runs locally on each executor.
 #'
+#' @importFrom SparkR sparkR.callJStatic sparkR.callJMethod
+#'
 #' @export
 setMethod("pipe",
           signature(ardd = "GenomicRDD",
@@ -185,8 +243,11 @@ setMethod("pipe",
 #' Sorts our genome aligned data by reference positions, with contigs ordered
 #' by index.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns a new, sorted RDD, of the implementing class type.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("sort",
           signature(ardd = "GenomicRDD"),
@@ -197,8 +258,11 @@ setMethod("sort",
 #' Sorts our genome aligned data by reference positions, with contigs ordered
 #' lexicographically.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns a new, sorted RDD, of the implementing class type.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("sortLexicographically",
           signature(ardd = "GenomicRDD"),
@@ -211,6 +275,8 @@ setMethod("sortLexicographically",
 #' @param ardd The RDD to convert into a dataframe.
 #' @return Returns a dataframe representing this RDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toDF",
           signature(ardd = "GenomicDataset"),
@@ -219,6 +285,7 @@ setMethod("toDF",
               new("SparkDataFrame", sdf, FALSE)
           })
 
+#' @importFrom SparkR sparkR.callJStatic
 setMethod("wrapTransformation",
           signature(ardd = "GenomicRDD",
                     tFn = "function"),
@@ -235,10 +302,13 @@ setMethod("wrapTransformation",
 #' Applies a function that transforms the underlying DataFrame into a new DataFrame
 #' using the Spark SQL API.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param tFn A function that transforms the underlying RDD as a DataFrame.
 #' @return A new RDD where the RDD of genomic data has been replaced, but the
 #'    metadata (sequence dictionary, and etc) is copied without modification.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("transform",
           signature(ardd = "GenomicRDD",
@@ -283,12 +353,15 @@ setMethod("destClassSuffix",
 #' Applies a function that transmutes the underlying DataFrame into a new RDD of a
 #' different type.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param tFn A function that transforms the underlying RDD as a DataFrame.
-#' @param convFn The name of the ADAM GenomicDatasetConversion class to use.
 #' @param destClass The destination class of this transmutation.
+#' @param convFn The name of the ADAM GenomicDatasetConversion class to use.
 #' @return A new RDD where the RDD of genomic data has been replaced, but the
 #'   metadata (sequence dictionary, and etc) is copied without modification.
 #'
+#' @importFrom SparkR sparkR.callJMethod sparkR.callJStatic
+#'
 #' @export
 setMethod("transmute",
           signature(ardd = "GenomicRDD",
@@ -326,9 +399,12 @@ setMethod("inferConversionFn",
 
 #' Convert this set of reads into fragments.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns a FragmentRDD where all reads have been grouped together by
 #' the original sequence fragment they come from.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toFragments",
           signature(ardd = "AlignmentRecordRDD"),
@@ -338,6 +414,7 @@ setMethod("toFragments",
 
 #' Saves this RDD to disk as a SAM/BAM/CRAM file.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath The path to save the file to.
 #' @param asType The type of file to save. Valid choices are SAM, BAM,
 #'   CRAM, and NA. If None, the file type is inferred from the extension.
@@ -345,6 +422,8 @@ setMethod("toFragments",
 #' @param asSingleFile Whether to save the file as a single merged
 #'   file or as shards.
 #'
+#' @importFrom SparkR sparkR.callJMethod sparkR.callJStatic
+#'
 #' @export
 setMethod("saveAsSam",
           signature(ardd = "AlignmentRecordRDD", filePath = "character"),
@@ -374,10 +453,13 @@ setMethod("saveAsSam",
 
 #' Converts this set of reads into a corresponding CoverageRDD.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param collapse Determines whether to merge adjacent coverage elements with
 #'                 the same score to a single coverage observation.
 #' @return Returns an RDD with observed coverage.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toCoverage",
           signature(ardd = "AlignmentRecordRDD"),
@@ -387,9 +469,12 @@ setMethod("toCoverage",
 
 #' Saves this RDD to disk, with the type identified by the extension.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath The path to save the file to.
 #' @param isSorted Whether the file is sorted or not.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("save",
           signature(ardd = "AlignmentRecordRDD", filePath = "character"),
@@ -399,9 +484,12 @@ setMethod("save",
 
 #' Cuts reads into _k_-mers, and then counts the occurrences of each _k_-mer.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param kmerLength The value of _k_ to use for cutting _k_-mers.
 #' @return Returns a DataFrame containing k-mer/count pairs.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("countKmers",
           signature(ardd = "AlignmentRecordRDD", kmerLength = "numeric"),
@@ -421,8 +509,11 @@ setMethod("countKmers",
 #' put at the end and sorted by read name. Contigs are ordered lexicographically
 #' by name.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return A new, sorted AlignmentRecordRDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("sortReadsByReferencePosition",
           signature(ardd = "AlignmentRecordRDD"),
@@ -436,8 +527,11 @@ setMethod("sortReadsByReferencePosition",
 #' put at the end and sorted by read name. Contigs are ordered by index that
 #' they are ordered in the sequence metadata.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return A new, sorted AlignmentRecordRDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("sortReadsByReferencePositionAndIndex",
           signature(ardd = "AlignmentRecordRDD"),
@@ -447,9 +541,12 @@ setMethod("sortReadsByReferencePositionAndIndex",
 
 #' Marks reads as possible fragment duplicates.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return A new RDD where reads have the duplicate read flag set. Duplicate
 #'          reads are NOT filtered out.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("markDuplicates",
           signature(ardd = "AlignmentRecordRDD"),
@@ -461,9 +558,13 @@ setMethod("markDuplicates",
 #'
 #' Uses a table of known SNPs to mask true variation during the recalibration
 #' process.
+#'
+#' @param ardd The RDD to apply this to.
 #' @param knownSnps A table of known SNPs to mask valid variants.
 #' @param validationStringency The stringency to apply towards validating BQSR.
 #'
+#' @importFrom SparkR sparkR.callJMethod sparkR.callJStatic
+#'
 #' @export
 setMethod("recalibrateBaseQualities",
           signature(ardd = "AlignmentRecordRDD", knownSnps = "VariantRDD", validationStringency = "character"),
@@ -472,91 +573,66 @@ setMethod("recalibrateBaseQualities",
               AlignmentRecordRDD(sparkR.callJMethod(ardd@jrdd, "recalibrateBaseQualities", knownSnps@jrdd, stringency))
           })
 
-#' Realigns indels using a concensus-based heuristic.
-#'
-#' Generates consensuses from reads.
+#' Realigns indels using a consensus-based heuristic.
 #'
+#' If no known indels are provided, generates consensuses from reads. Else,
+#' generates consensuses from previously seen variants.
+#' 
+#' @param ardd The RDD to apply this to.
 #' @param isSorted If the input data is sorted, setting this parameter to true
 #'   avoids a second sort.
-#' @param int maxIndelSize The size of the largest indel to use for realignment.
+#' @param maxIndelSize The size of the largest indel to use for realignment.
 #' @param maxConsensusNumber The maximum number of consensus sequences to
 #'   realign against per target region.
 #' @param lodThreshold Log-odds threshold to use when realigning; realignments
 #'   are only finalized if the log-odds threshold is exceeded.
 #' @param maxTargetSize The maximum width of a single target region for
 #'   realignment.
+#' @param knownIndels An RDD of previously called INDEL variants.
 #' @return Returns an RDD of mapped reads which have been realigned.
 #'
+#' @importFrom SparkR sparkR.callJMethod sparkR.callJStatic
+#'
 #' @export
 setMethod("realignIndels",
           signature(ardd = "AlignmentRecordRDD"),
           function(ardd, isSorted = FALSE, maxIndelSize = 500,
                    maxConsensusNumber = 30, lodThreshold = 5.0,
-                   maxTargetSize = 3000) {
-              consensusModel <- sparkR.callJStatic("org.bdgenomics.adam.algorithms.consensus.ConsensusGenerator",
-                                                   "fromReads")
-              AlignmentRecordRDD(sparkR.callJMethod(ardd@jrdd, "realignIndels",
-                                                    consensusModel,
-                                                    isSorted,
-                                                    maxIndelSize,
-                                                    maxConsensusNumber,
-                                                    lodThreshold,
-                                                    maxTargetSize))
-          })
+                   maxTargetSize = 3000,
+                   knownIndels = NA) {
+
+              if (is.na(knownIndels)) {
+                  consensusModel <- sparkR.callJStatic("org.bdgenomics.adam.algorithms.consensus.ConsensusGenerator",
+                                                       "fromKnowns", knownIndels@jrdd)
+                  AlignmentRecordRDD(sparkR.callJMethod(ardd@jrdd, "realignIndels",
+                                                        consensusModel,
+                                                        isSorted,
+                                                        maxIndelSize,
+                                                        maxConsensusNumber,
+                                                        lodThreshold,
+                                                        maxTargetSize))
 
-#' Realigns indels using a concensus-based heuristic.
-#'
-#' Generates consensuses from previously seen variants.
-#'
-#' @param knownIndels An RDD of previously called INDEL variants.
-#' @param isSorted If the input data is sorted, setting this parameter to true
-#'   avoids a second sort.
-#' @param int maxIndelSize The size of the largest indel to use for realignment.
-#' @param maxConsensusNumber The maximum number of consensus sequences to
-#'   realign against per target region.
-#' @param lodThreshold Log-odds threshold to use when realigning; realignments
-#'   are only finalized if the log-odds threshold is exceeded.
-#' @param maxTargetSize The maximum width of a single target region for
-#'   realignment.
-#' @return Returns an RDD of mapped reads which have been realigned.
-#'
-#' @export
-setMethod("realignIndels",
-          signature(ardd = "AlignmentRecordRDD", knownIndels = "VariantRDD"),
-          function(ardd, knownIndels, isSorted = FALSE, maxIndelSize = 500,
-                   maxConsensusNumber = 30, lodThreshold = 5.0,
-                   maxTargetSize = 3000) {
-              consensusModel <- sparkR.callJStatic("org.bdgenomics.adam.algorithms.consensus.ConsensusGenerator",
-                                                   "fromKnowns", knownIndels@jrdd)
-              AlignmentRecordRDD(sparkR.callJMethod(ardd@jrdd, "realignIndels",
-                                                    consensusModel,
-                                                    isSorted,
-                                                    maxIndelSize,
-                                                    maxConsensusNumber,
-                                                    lodThreshold,
-                                                    maxTargetSize))
-          })
-
-setMethod("replaceRdd",
-          signature(ardd = "CoverageRDD",
-                    rdd = "jobj"),
-          function(ardd, rdd) {
-              CoverageRDD(rdd)
-          })
-
-setMethod("inferConversionFn",
-          signature(ardd = "CoverageRDD",
-                    destClass = "character"),
-          function(ardd, destClass) {
-              paste0("org.bdgenomics.adam.api.java.CoverageTo",
-                     destClassSuffix(destClass))
+              } else {
+                  consensusModel <- sparkR.callJStatic("org.bdgenomics.adam.algorithms.consensus.ConsensusGenerator",
+                                                       "fromReads")
+                  AlignmentRecordRDD(sparkR.callJMethod(ardd@jrdd, "realignIndels",
+                                                        consensusModel,
+                                                        isSorted,
+                                                        maxIndelSize,
+                                                        maxConsensusNumber,
+                                                        lodThreshold,
+                                                        maxTargetSize))
+              }
           })
 
 #' Saves coverage as a feature file.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath The location to write the output.
 #' @param asSingleFile If true, merges the sharded output into a single file.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("save",
           signature(ardd = "CoverageRDD", filePath = "character"),
@@ -571,8 +647,11 @@ setMethod("save",
 #' 3.0) and Coverage("chr1", 10, 20, 3.0) would be merged into one record
 #' Coverage("chr1", 1, 20, 3.0).
 #'
+#' @param ardd The RDD to apply this to.
 #' @return An RDD with merged tuples of adjacent sites with same coverage.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("collapse", signature(ardd = "CoverageRDD"),
           function(ardd) {
@@ -581,8 +660,11 @@ setMethod("collapse", signature(ardd = "CoverageRDD"),
 
 #' Converts CoverageRDD to FeatureRDD.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns a FeatureRDD from a CoverageRDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toFeatures", signature(ardd = "CoverageRDD"),
           function(ardd) {
@@ -595,9 +677,12 @@ setMethod("toFeatures", signature(ardd = "CoverageRDD"),
 #' bin together ReferenceRegions of equal size. The coverage of each bin is the
 #' coverage of the first base pair in that bin.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param bpPerBin Number of bases to combine to one bin.
 #' @return Returns a sparsified CoverageRDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("coverage", signature(ardd = "CoverageRDD"),
           function(ardd, bpPerBin = 1) {
@@ -610,9 +695,14 @@ setMethod("coverage", signature(ardd = "CoverageRDD"),
 #' bin together ReferenceRegions of equal size. The coverage of each bin is the
 #' average coverage of the bases in that bin.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param bpPerBin Number of bases to combine to one bin.
 #' @return Returns a sparsified CoverageRDD.
 #'
+#' @rdname CoverageRDD
+#'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("aggregatedCoverage", signature(ardd = "CoverageRDD"),
           function(ardd, bpPerBin = 1) {
@@ -623,8 +713,11 @@ setMethod("aggregatedCoverage", signature(ardd = "CoverageRDD"),
 #'
 #' The opposite operation of collapse.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return New CoverageRDD of flattened coverage.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("flatten", signature(ardd = "CoverageRDD"),
           function(ardd) {
@@ -653,11 +746,14 @@ setMethod("replaceRdd",
 #' these match, we fall back to Parquet. These files are written as sharded text
 #' files, which can be merged by passing asSingleFile = True.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath The location to write the output.
 #' @param asSingleFile If true, merges the sharded output into a single file.
 #' @param disableFastConcat If asSingleFile is true, disables the use of the
 #'  fast concatenation engine for saving to HDFS.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("save",
           signature(ardd = "FeatureRDD", filePath = "character"),
@@ -669,8 +765,11 @@ setMethod("save",
 
 #' Converts the FeatureRDD to a CoverageRDD.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns a new CoverageRDD.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toCoverage", signature(ardd = "FeatureRDD"),
           function(ardd) {
@@ -694,8 +793,11 @@ setMethod("replaceRdd",
 
 #' Splits up the reads in a Fragment, and creates a new RDD.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns this RDD converted back to reads.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toReads", signature(ardd = "FragmentRDD"),
           function(ardd) {
@@ -704,9 +806,12 @@ setMethod("toReads", signature(ardd = "FragmentRDD"),
 
 #' Marks reads as possible fragment duplicates.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return A new RDD where reads have the duplicate read flag set. Duplicate
 #'   reads are NOT filtered out.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("markDuplicates", signature(ardd = "FragmentRDD"),
           function(ardd) {
@@ -715,8 +820,11 @@ setMethod("markDuplicates", signature(ardd = "FragmentRDD"),
 
 #' Saves fragments to Parquet.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath Path to save fragments to.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("save", signature(ardd = "FragmentRDD", filePath = "character"),
           function(ardd, filePath) {
@@ -740,8 +848,11 @@ setMethod("replaceRdd",
 
 #' Saves this RDD of genotypes to disk as Parquet.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath Path to save file to.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("saveAsParquet", signature(ardd = "GenotypeRDD", filePath = "character"),
           function(ardd, filePath) {
@@ -750,8 +861,11 @@ setMethod("saveAsParquet", signature(ardd = "GenotypeRDD", filePath = "character
 
 #' Converts this RDD of Genotypes to VariantContexts.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns this RDD of Genotypes as VariantContexts.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toVariantContexts", signature(ardd = "GenotypeRDD"),
           function(ardd) {
@@ -778,8 +892,11 @@ setMethod("replaceRdd",
 #' If filename ends in .fa or .fasta, saves as Fasta. If not, saves fragments to
 #' Parquet. Defaults to 60 character line length, if saving as FASTA.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath Path to save to.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("save", signature(ardd = "NucleotideContigFragmentRDD", filePath = "character"),
           function(ardd, filePath) {
@@ -789,10 +906,13 @@ setMethod("save", signature(ardd = "NucleotideContigFragmentRDD", filePath = "ch
 #' For all adjacent records in the RDD, we extend the records so that the
 #' adjacent records now overlap by _n_ bases, where _n_ is the flank length.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param flankLength The length to extend adjacent records by.
 #' @return Returns the RDD, with all adjacent fragments extended with flanking
 #'   sequence.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("flankAdjacentFragments",
           signature(ardd = "NucleotideContigFragmentRDD", flankLength = "numeric"),
@@ -819,8 +939,11 @@ setMethod("replaceRdd",
 
 #' Saves this RDD of variants to disk as Parquet.
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath Path to save file to.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("saveAsParquet", signature(ardd = "VariantRDD", filePath = "character"),
           function(ardd, filePath) {
@@ -829,8 +952,11 @@ setMethod("saveAsParquet", signature(ardd = "VariantRDD", filePath = "character"
 
 #' Converts this RDD of Variants to VariantContexts.
 #'
+#' @param ardd The RDD to apply this to.
 #' @return Returns this RDD of Variants as VariantContexts.
 #'
+#' @importFrom SparkR sparkR.callJMethod
+#'
 #' @export
 setMethod("toVariantContexts", signature(ardd = "VariantRDD"),
           function(ardd) {
@@ -846,15 +972,18 @@ setMethod("replaceRdd",
 
 #' Saves this RDD of variant contexts to disk as VCF
 #'
+#' @param ardd The RDD to apply this to.
 #' @param filePath Path to save VCF to.
 #' @param asSingleFile If true, saves the output as a single file
 #'   by merging the sharded output after saving.
 #' @param deferMerging If true, saves the output as prepped for merging
 #'   into a single file, but does not merge.
 #' @param stringency The stringency to use when writing the VCF.
-#' @param disableFastConcat: If asSingleFile is true, disables the use
+#' @param disableFastConcat If asSingleFile is true, disables the use
 #'   of the fast concatenation engine for saving to HDFS.
 #'
+#' @importFrom SparkR sparkR.callJMethod sparkR.callJStatic
+#'
 #' @export
 setMethod("saveAsVcf", signature(ardd = "VariantContextRDD", filePath = "character"),
           function(ardd,
diff --git a/adam-r/bdgenomics.adam/tests/testthat/test_adamContext.R b/adam-r/bdgenomics.adam/tests/testthat/test_adamContext.R
index f5cdcc6aa8..8582117c02 100644
--- a/adam-r/bdgenomics.adam/tests/testthat/test_adamContext.R
+++ b/adam-r/bdgenomics.adam/tests/testthat/test_adamContext.R
@@ -16,11 +16,11 @@
 # limitations under the License.
 #
 library(bdgenomics.adam)
+library(SparkR)
 
 context("basic ADAM context functions")
 
-sc <- sparkR.session()
-ac <- ADAMContext(sc)
+ac <- createADAMContext()
 
 test_that("load reads", {
     reads <- loadAlignments(ac, resourceFile("small.sam"))
diff --git a/adam-r/bdgenomics.adam/tests/testthat/test_alignmentRecordRdd.R b/adam-r/bdgenomics.adam/tests/testthat/test_alignmentRecordRdd.R
index ec1eb70410..ae91240244 100644
--- a/adam-r/bdgenomics.adam/tests/testthat/test_alignmentRecordRdd.R
+++ b/adam-r/bdgenomics.adam/tests/testthat/test_alignmentRecordRdd.R
@@ -19,8 +19,7 @@ library(bdgenomics.adam)
 
 context("manipulating alignmentrecords")
 
-sc <- sparkR.session()
-ac <- ADAMContext(sc)
+ac <- createADAMContext()
 
 test_that("save sorted sam", {
 
diff --git a/adam-r/bdgenomics.adam/tests/testthat/test_featureRdd.R b/adam-r/bdgenomics.adam/tests/testthat/test_featureRdd.R
index 27b789a7a7..ba10e6103d 100644
--- a/adam-r/bdgenomics.adam/tests/testthat/test_featureRdd.R
+++ b/adam-r/bdgenomics.adam/tests/testthat/test_featureRdd.R
@@ -19,8 +19,7 @@ library(bdgenomics.adam)
 
 context("manipulating features")
 
-sc <- sparkR.session()
-ac <- ADAMContext(sc)
+ac <- createADAMContext()
 
 test_that("round trip gtf", {
     testFile <- resourceFile("Homo_sapiens.GRCh37.75.trun20.gtf")
diff --git a/adam-r/bdgenomics.adam/tests/testthat/test_genotypeRdd.R b/adam-r/bdgenomics.adam/tests/testthat/test_genotypeRdd.R
index 04bf83e186..abb29326d8 100644
--- a/adam-r/bdgenomics.adam/tests/testthat/test_genotypeRdd.R
+++ b/adam-r/bdgenomics.adam/tests/testthat/test_genotypeRdd.R
@@ -19,8 +19,7 @@ library(bdgenomics.adam)
 
 context("manipulating genotypes")
 
-sc <- sparkR.session()
-ac <- ADAMContext(sc)
+ac <- createADAMContext()
 
 test_that("round trip vcf", {
     testFile <- resourceFile("small.vcf")
diff --git a/adam-r/bdgenomics.adam/tests/testthat/test_variantRdd.R b/adam-r/bdgenomics.adam/tests/testthat/test_variantRdd.R
index 49893b07dd..e9fb59d6f7 100644
--- a/adam-r/bdgenomics.adam/tests/testthat/test_variantRdd.R
+++ b/adam-r/bdgenomics.adam/tests/testthat/test_variantRdd.R
@@ -19,8 +19,7 @@ library(bdgenomics.adam)
 
 context("manipulating variants")
 
-sc <- sparkR.session()
-ac <- ADAMContext(sc)
+ac <- createADAMContext()
 
 test_that("round trip vcf", {
     testFile <- resourceFile("small.vcf")
diff --git a/scripts/release/release.sh b/scripts/release/release.sh
index bb5443bc69..b07b4a4124 100755
--- a/scripts/release/release.sh
+++ b/scripts/release/release.sh
@@ -129,6 +129,18 @@ rm -rf release-venv
 
 popd
 
+# build R tarball
+#
+# !!!!!
+# NOTE:
+# !!!!!
+#
+# We will not be pushing to CRAN until SparkR is reinstated in CRAN. Until then,
+# this tarball will need to be manually attached to the github releases page
+pushd adam-r
+R CMD build bdgenomics.adam
+popd
+
 if [ $? != 0 ]; then
   echo "Releasing bdgenomics.adam to PyPi failed."
   exit 1