version 0.1.0

DESCRIPTION

@@ -0,0 +1,35 @@
+Package: theseus
+Type: Package
+Title: Analysis and Visualization Tools for Microbial Community Data
+Version: 0.1.0
+Authors@R: c(
+    person("Jacob", "Price", email = "jrp355@drexel.edu", role = c("aut")),
+    person("Stephen", "Woloszynek", email = "sw424@drexel.edu", role = c("cre","aut"))
+    )
+Description: An approach to the visualization, analysis, and interpretation of 
+  (microbial) community composition data, especially those originating from 
+  amplicon sequencing. Analysis techniques include constrained and unconstrained
+  ordination and visualizing taxonomic abundances and spatial patterns, among 
+  others. Methods intended to assist bioinformaticians and ecologists in 
+  selecting read trimming by quality scores and preprocessing/denoising of 
+  datasets are also provided. 
+License: MIT + file LICENSE
+URL: http://github.com/EESI/theseus
+BugReports: http://github.com/EESI/theseus/issues
+Encoding: UTF-8
+LazyData: TRUE
+Depends: R (>= 3.3.0)
+Imports: dplyr, ggplot2 (>= 2.2.1), gridExtra, magrittr (>= 1.5),
+        parallel (>= 3.4.1), phyloseq (>= 1.20.0), ShortRead (>=
+        1.34.0), splancs, tidyverse (>= 1.1.1), tidyr, vegan (>=
+        2.3.5), viridis (>= 0.4.0)
+Suggests: covr, knitr, rmarkdown, testthat
+VignetteBuilder: knitr
+RoxygenNote: 6.0.1
+NeedsCompilation: no
+Packaged: 2017-12-20 14:50:53 UTC; sw1
+Author: Jacob Price [aut],
+  Stephen Woloszynek [cre, aut]
+Maintainer: Stephen Woloszynek <sw424@drexel.edu>
+Repository: CRAN
+Date/Publication: 2017-12-20 16:30:53 UTC

LICENSE

@@ -0,0 +1,2 @@
+YEAR: 2017
+COPYRIGHT HOLDER: Jacob R. Price

MD5

@@ -0,0 +1,50 @@
+59d3a39288ceb8f0fe5dd52ae995b991 *DESCRIPTION
+2c2363b99fe377f5f12d0facb102d9a1 *LICENSE
+059bde17125ec613230f3a95977149b3 *NAMESPACE
+dcbd9db04464280430451e1a335ee87a *NEWS.md
+4dc3288130c6430b7c7d6cfa28f55f5d *R/cohort_relabund.R
+2dd5ac84ff08809c1e2ecae3326c4864 *R/constord.R
+384e3e8125444284ea8be7ce13fe774a *R/data_WWTP_Impact.R
+831076f68f2c61a8982ee6cc4d7243a0 *R/data_seqtab.2vs3.R
+cdff655314a9907bbdf2df69be6dd544 *R/data_seqtab.R
+562eb616187c3f57375243f05e1c0164 *R/envtoverlay.R
+165f199fbf2be26c1a4a31de85c1b75f *R/misc.R
+eb37a34591b06316d60637c3da5df788 *R/prev.R
+f4b666deb166165658915f89c0d732b6 *R/pstoveg_otu.R
+de8a1b4705fd19b52dcf13dbff2193aa *R/pstoveg_sample.R
+64188d7b2b6bd858219516345a5cd8e2 *R/pstoveg_sd.R
+1d1c05e8b8a5cc89e5d05e1bfac6ab15 *R/qualcontour.R
+529ef7af731485e8c8e65f99e2725d90 *R/utils.R
+2628fa47d7d9f912c46ed4f79913bcc3 *README.md
+00b39ee1382469cd71c77abace310140 *build/vignette.rds
+46009cefc529bb6afbe1999556a04f56 *data/WWTP_Impact.RData
+84ebf41ff653529cd2789e881853736c *data/sigtab.2vs3.RData
+91e2e5c1c22a91e8816493f87da114b9 *data/sigtab.RData
+00aaad90eb2ceb966aff6873b999d51f *inst/CITATION
+e4f10a71145707d13b62ea38621f905b *inst/doc/general_usage.R
+99c285aa3354c367077ea07c16f118e0 *inst/doc/general_usage.Rmd
+1c6cbd42950ae779b94616767a39dd34 *inst/doc/general_usage.html
+4ce15504a6287a760e5522e4715df85f *inst/testdata/sample1_R1.fastq.gz
+7957b1564084964551ea64d86a108406 *inst/testdata/sample1_R2.fastq.gz
+0f5d5c73d2f0dba0dfb77ef289339813 *inst/testdata/sample2_R1.fastq.gz
+92ff41e67c790e3f7f83d852a400f5d7 *inst/testdata/sample2_R2.fastq.gz
+a18d75d4e199b6f1db9479fa4075d005 *man/WWTP_Impact.Rd
+9e8ea1081830c57f6d08de5b70b66d18 *man/cohort_relabund.Rd
+afb157d72b93887a17d5fe4bb351e59e *man/constord.Rd
+5434a1b83f6dba6ffeb8b6c8ad7b4c9b *man/envtoverlay.Rd
+b553be0c9adc706380507835afff3c35 *man/prev.Rd
+65161921a3d904cad105798a4576ce76 *man/pstoveg_otu.Rd
+601b6e6c6dc267108d0f4a3e07ed9c80 *man/pstoveg_sample.Rd
+599a43291ca24f9b86836e65014da5e9 *man/pstoveg_sd.Rd
+3f621de6d164eb97fc6532811dbfeec1 *man/qualcontour.Rd
+01dc6ddd7c976a13d72a587f3938f483 *man/reexports.Rd
+e2761cadc94f71a915d9a9576c05bbcf *man/sigtab.2vs3.Rd
+d7c077a6ee9a778a5c254940e68a05bb *man/sigtab.Rd
+d970962f446e6b4b66c98511ba071afe *tests/testthat.R
+1052aea0f05b2dd6003f165039ce704a *tests/testthat/test-helpers.R
+99c285aa3354c367077ea07c16f118e0 *vignettes/general_usage.Rmd
+b5f516aa7bf9c9647a03dddfb40d9f59 *vignettes/p.co.png
+788aa3636cea9d859848c1f75118c7de *vignettes/p.cra.png
+6b9eac634279acf6f0ae2ad9eaafba8c *vignettes/p.eo.png
+aeda3287e002a57877bf07b0b50e919e *vignettes/p.prev.png
+91768f498ae6724a6ca7675d08db386c *vignettes/p.qc.png

NAMESPACE

@@ -0,0 +1,24 @@
+# Generated by roxygen2: do not edit by hand
+
+export("%>%")
+export("otu_table<-")
+export("sample_data<-")
+export("tax_table<-")
+export(cohort_relabund)
+export(constord)
+export(envtoverlay)
+export(prev)
+export(pstoveg_otu)
+export(pstoveg_sample)
+export(pstoveg_sd)
+export(qualcontour)
+import(ggplot2)
+importFrom(magrittr,"%>%")
+importFrom(phyloseq,"otu_table<-")
+importFrom(phyloseq,"sample_data<-")
+importFrom(phyloseq,"tax_table<-")
+importFrom(phyloseq,otu_table)
+importFrom(phyloseq,plot_bar)
+importFrom(phyloseq,sample_data)
+importFrom(phyloseq,tax_table)
+importFrom(phyloseq,transform_sample_counts)

NEWS.md

@@ -0,0 +1,3 @@
+# theseus 0.1.0
+
+* First release.

R/cohort_relabund.R

@@ -0,0 +1,140 @@
+#' @import ggplot2
+#' @importFrom phyloseq transform_sample_counts plot_bar
+NULL
+
+#' Creates a relative abundance cohort plot
+#'
+#' This function plots the relative abundance of taxa within a phyloseq
+#' object 'PS' according to thier pre-determined cohort memberships.
+#' See Details for more information.
+#'
+#' @param PS (required) A phyloseq object.
+#' @param xvar (required) Variable in sample_data(PS) to be displayed on the
+#'   x-axis. Defaults to 'SampleID'.
+#' @param taxfill Taxonomic level to display in plot. Defaults to 'Phylum'.
+#' @param comp1 (required) First comparison (hence 'comp1') object of
+#'   'DESeqResults' class, or a dataframe with similar structure. In the case
+#'   that a 'DESeqResults' class object is not being used, the object must
+#'   contain a 'log2FoldChange' vector/column. Row names must be a subset (but
+#'   not necessarily a proper subset) of taxa_names(PS). See Details for more
+#'   information.
+#' @param comp2 (required) Second comparison (hence 'comp2') object. Refer to
+#'   documentation for 'comp1' for remaining details.
+#' @param comp1lab Labels for comparison 1. Defaults to c('Decreased Comp1',
+#'   'No Change Comp1','Increased Comp1').
+#' @param comp2lab Labels for comparison 2. Defaults to c('Decreased Comp2',
+#'   'No Change Comp2','Increased Comp2').
+#' @param justdata Return only the data table (no plot). Defaults to FALSE.
+#' @param PSisRelAbund Does the PS object contain compositional (relative
+#'   abundance) taxa counts? Defaults to FALSE.
+#'
+#' @details
+#' \subsection{General}{
+#' The results from a single pairwise comparison, such as pre- and
+#'   post-treatment, carried out with the DESeq2 package can be plotted or read
+#'   and interpreted in tabular form with relative ease. When two pairwise
+#'   comparisons are being performed, interpreting the results becomes more
+#'   difficult. This function is intended to assist with interpreting the
+#'   results from multiple differential abundance analyses carried out with the
+#'   DESeq2 r-package. This function takes a phyloseq object ('PS') and two
+#'   'DESeqResults' objects ('comp1', 'comp2') and plots the relative abundance
+#'   of taxa within 'PS', partitioning the taxa according to their membership
+#'   to one the 9 possible cohort combinations determined by their values
+#'   specified within 'comp1' and 'comp2'.
+#' }
+#' \subsection{Approach}{
+#' The \code{\link[DESeq2]{DESeq}} function carries out differential abundance
+#' testing and produces a 'DESeqDataSet' object. The
+#' \code{\link[DESeq2]{results}} function can be used to access the results and
+#' create a 'DESeqResults' object, which is a subclass of DataFrame. Note that
+#' the 'alpha' parameter for \code{\link[DESeq2]{results}} can be used to
+#' specify the significance level of the test being performed. NOTE: Testing
+#' with DESeq2 must be carried out, and the non-significant taxa should be
+#' removed from the 'comp1' and 'comp2' objects before using this function.
+#' Using the log2FoldChange columns in 'comp1' and 'comp2' this function
+#' identifies which taxa decrease, do not change, or increase over course of
+#' both comparisons. Because there are three options for both comparisons
+#' there are 3^2=9 possible combinations, or cohorts, which an OTU may fall
+#' into. These cohort assignments are used when plotting the relative abundance
+#' plot.
+#' }
+#'
+#'
+#' @return A ggplot object.
+#'
+#' @seealso \code{\link[phyloseq]{plot_bar}}
+#'   \code{\link[phyloseq]{transform_sample_counts}}
+#'   \code{\link[DESeq2]{DESeq}} \code{\link[DESeq2]{results}}
+#'   \code{\link[DESeq2]{DESeqDataSet}}
+#'
+#' @examples
+#' \dontrun{
+#' cohort_relabund(
+#'   PS=prune_samples(sample_data(WWTP_Impact)$site %in% c(1,2,3,4),
+#'                    WWTP_Impact),
+#'   comp1=sigtab,
+#'   comp2=sigtab.2vs3,
+#'   comp1lab=c('Decreased at Effluent',
+#'              'No change at Effluent',
+#'              'Increased at effluent'),
+#'   comp2lab=c('Decreased btwn plants',
+#'              'No change btwn plants',
+#'              'Increased btwn plants'))
+#' }
+#'
+#' @export
+
+cohort_relabund <- function(PS,xvar='SampleID',taxfill='Phylum',comp1,comp2,comp1lab=c('Decreased Comp1','No Change Comp1','Increased Comp1'),comp2lab=c('Decreased Comp2','No Change Comp2','Increased Comp2'), justdata=FALSE, PSisRelAbund=FALSE){
+
+	ls.comp1.up <- rownames(comp1[comp1$log2FoldChange>0,])
+	ls.comp1.down <- rownames(comp1[comp1$log2FoldChange<0,])
+
+	ls.comp2.up <- rownames(comp2[comp2$log2FoldChange>0,])
+	ls.comp2.down <- rownames(comp2[comp2$log2FoldChange<0,])
+
+	if (PSisRelAbund==FALSE){
+		psra <- transform_sample_counts(PS,function(x) {x / sum(x)})
+	} else {
+		psra <- PS
+	}
+
+	bp <- plot_bar(psra, x=xvar, fill=taxfill)
+	bpdat <- bp$data
+
+	#bpdat$comp1 <- comp1lab[2]
+	#bpdat$comp1[bpdat$OTU %in% ls.comp1.up]
+	bpdat$comp1cat[bpdat$OTU %in% ls.comp1.down] <- comp1lab[1]
+	bpdat$comp1cat[!bpdat$OTU %in% c(ls.comp1.down,ls.comp1.up)] <- comp1lab[2]
+	bpdat$comp1cat[bpdat$OTU %in% ls.comp1.up] <- comp1lab[3]
+	bpdat$comp1cat <-
+		factor(bpdat$comp1cat,
+					 levels=c(comp1lab[1], comp1lab[2], comp1lab[3]), ordered=TRUE)
+
+
+	bpdat$comp2cat[bpdat$OTU %in% ls.comp2.down] <- comp2lab[1]
+	bpdat$comp2cat[!bpdat$OTU %in% c(ls.comp2.down,ls.comp2.up)] <- comp2lab[2]
+	bpdat$comp2cat[bpdat$OTU %in% ls.comp2.up] <- comp2lab[3]
+	bpdat$comp2cat <-
+		factor(bpdat$comp2cat,
+					 levels=c(comp2lab[1], comp2lab[2], comp2lab[3]), ordered=TRUE)
+
+	if (justdata==TRUE){
+		bpdat
+	} else {
+
+	p1 <-
+	ggplot(bpdat, aes_string(x=xvar, y='Abundance', fill=taxfill)) +
+		geom_bar(stat='identity', position='stack') +
+		theme(axis.text.x = element_text(angle=90, vjust=0.5, hjust=0)) +
+		facet_grid(comp2cat~comp1cat) +
+		theme(legend.position = 'bottom') +
+		#theme(legend.title=element_blank()) +
+		theme(axis.title.x=element_blank()) #+
+		#geom_vline(xintercept=35-5-0.5, linetype='solid', alpha=1.0, size=1.5)
+		#ylab('Relative Abundance (Phylum)')
+
+	p1
+	}
+}
+
+