Rename protcomp() to human.aa()

DESCRIPTION

@@ -1,6 +1,6 @@
-Date: 2024-03-02
+Date: 2024-03-03
 Package: canprot
-Version: 1.1.2-28
+Version: 1.1.2-29
 Title: Chemical Analysis of Proteins
 Authors@R: c(
     person("Jeffrey", "Dick", email = "[email protected]", role = c("aut", "cre"),

NAMESPACE

@@ -1,7 +1,7 @@
 # Exported variables
 export(
 # 20170612 (canprot 0.1.0)
-"protcomp", "CLES",
+"human.aa", "CLES",
 # metrics added 20191005 - 20191027
 "Zc", "nH2O", "GRAVY", "pI",
 # 20200204 used in JMDplots package

R/cplab.R

@@ -3,28 +3,45 @@
 # Moved from diffplot.R and changed expression() to quote() 20230617
 
 cplab <- list(
-  nH2O = quote(italic(n)[H[2]*O]),
-  DnH2O = quote(Delta*italic(n)[H[2]*O]),
-  nO2 = quote(italic(n)[O[2]]),
-  DnO2 = quote(Delta*italic(n)[O[2]]),
   Zc = quote(italic(Z)[C]),
   DZc = quote(Delta*italic(Z)[C]),
+  nO2 = quote(italic(n)[O[2]]),
+  DnO2 = quote(Delta*italic(n)[O[2]]),
+  nH2O = quote(italic(n)[H[2]*O]),
+  DnH2O = quote(Delta*italic(n)[H[2]*O]),
+  logfO2 = quote(log~italic("f")[O[2]]),
+  logaH2O = quote(log~italic("a")[H[2]*O]),
   nC = quote(italic(n)[C] * "/AA"),
+  nH = quote(italic(n)[H] * "/AA"),
   nN = quote(italic(n)[N] * "/AA"),
+  nO = quote(italic(n)[O] * "/AA"),
   nS = quote(italic(n)[S] * "/AA"),
   DnC = quote(Delta*italic(n)[C] * "/AA"),
+  DnH = quote(Delta*italic(n)[H] * "/AA"),
   DnN = quote(Delta*italic(n)[N] * "/AA"),
+  DnO = quote(Delta*italic(n)[O] * "/AA"),
   DnS = quote(Delta*italic(n)[S] * "/AA"),
   V0 = quote("Volume per residue (cm" ^ 3 ~ "mol" ^ -1 * ")"),
   pV0 = quote("Volume per protein (cm" ^ 3 ~ "mol" ^ -1 * ")"),
-  DV0 = quote(list(Delta * italic("V") * degree, "cm" ^ 3 ~ "mol" ^ -1)),
+  DV0 = quote(Delta * "V per residue (cm" ^ 3 ~ "mol" ^ -1 * ")"),
+  DV0 = quote(Delta * "V per protein (cm" ^ 3 ~ "mol" ^ -1 * ")"),
   nAA = quote(italic(n)[AA]),
   DnAA = quote(Delta*italic(n)[AA]),
   GRAVY = "GRAVY",
   DGRAVY = quote(Delta*"GRAVY"),
   pI = "pI",
   DpI = quote(Delta*"pI"),
   MW = quote("MW per residue"),
-  DMW = quote(Delta * "MW"),
-  plength = "Protein length"
+  DMW = quote(Delta * "MW per residue"),
+  pMW = quote("MW per protein"),
+  DpMW = quote(Delta * "MW per protein"),
+  plength = "Protein length",
+  H_C = "H/C",
+  HC = "H/C",
+  N_C = "N/C",
+  NC = "N/C",
+  O_C = "O/C",
+  OC = "O/C",
+  S_C = "S/C",
+  SC = "S/C"
 )

R/human.aa.R

@@ -0,0 +1,36 @@
+# canprot/human.aa.R
+# Get amino acid compositions for human proteins from UniProt IDs
+# 20160705 jmd
+
+human.aa <- function(uniprot = NULL, aa_file = NULL, stop.if.missing = FALSE, warn.if.duplicated = FALSE) {
+  # Get amino acid compositions of human proteins
+  aa <- get("human_aa", canprot)
+  # Add amino acid compositions from external file if specified
+  if(!is.null(aa_file)) {
+    aa_dat <- read.csv(aa_file, as.is=TRUE)
+    print(paste("human.aa: adding", nrow(aa_dat), "proteins from", aa_file))
+    aa <- rbind(aa_dat, aa)
+  }
+  if(is.null(uniprot)) {
+    stop("'uniprot' is NULL")
+  } else {
+    # Find the proteins listed in 'uniprot' - first look at the ID after the | separator
+    alluni <- sapply(strsplit(aa$protein, "|", fixed = TRUE), "[", 2)
+    # If that is NA (i.e. no | separator is present) use the entire string
+    ina <- is.na(alluni)
+    alluni[ina] <- aa$protein[ina]
+    iuni <- match(uniprot, alluni)
+    if(stop.if.missing) {
+      # Stop with error if any IDs are not found
+      if(any(is.na(iuni))) stop(paste("uniprot IDs not found:", paste(uniprot[is.na(iuni)], collapse = " ")))
+    }
+    if(warn.if.duplicated) {
+      # Warn if any IDs are duplicated
+      if(any(duplicated(iuni))) warning(paste("some uniprot IDs are duplicated:",
+        paste(uniprot[duplicated(iuni)], collapse=" ")), immediate. = TRUE)
+    }
+    aa <- aa[iuni, ]
+  }
+  # Return amino acid compositions
+  aa
+}

R/metrics.R

@@ -123,12 +123,14 @@ pI <- function(AAcomp, terminal_H2O = 1, ...) {
     iZ <- match(names(AA)[isZ], dimnames(Ztab)[[2]])
     # Calculate the total charge as a function of pH
     # ... the "else" is in case we have a data frame (used when first writing this function)
-    if(is.numeric(AA)) Ztot <- Ztab[, iZ] %*% AA[isZ]
-    else Ztot <- Ztab[, iZ] %*% as.matrix(t(AA[, isZ]))
+    if(is.numeric(AA)) Ztot <- Ztab[, iZ] %*% AA[isZ] else
+    Ztot <- Ztab[, iZ] %*% as.matrix(t(AA[, isZ]))
     # Find pH where charge is closest to zero
     # (absolute charge is minimized)
     ipH <- which.min(abs(Ztot))
-    Ztab[ipH, 1]
+    pI <- Ztab[ipH, 1]
+    if(length(pI) == 0) pI <- NA
+    pI
   }
   # Number of N- and C-terminal groups
   Nterm <- Cterm <- terminal_H2O

inst/NEWS

@@ -1,5 +1,5 @@
-CHANGES IN canprot 1.1.2-27 (2024-03-01)
----------------------------------------
+CHANGES IN canprot 1.1.2-29 (2024-03-03)
+----------------------------------------
 
 - Objects in cplab are now `quote()`-ed instead of `expression()`s for easier
   handling by other functions.
@@ -26,6 +26,10 @@ CHANGES IN canprot 1.1.2-27 (2024-03-01)
 
 - Add add.cld() (add compact letter display to boxplots).
 
+- Move chem16S::calc_metrics() to calc.metrics().
+
+- Rename protcomp() to human.aa().
+
 CHANGES IN canprot 1.1.2 (2022-01-17)
 -------------------------------------
 

man/human.Rd

@@ -7,7 +7,7 @@
 \alias{human_additional}
 \alias{human_extra}
 \alias{uniprot_updates}
-\title{Amino Acid Compositions of Human Proteins}
+\title{Data for Amino Acid Compositions of Human Proteins}
 \description{
   Data for amino acid compositions of proteins and conversion from old to new UniProt IDs.
 }
@@ -43,7 +43,7 @@ The \code{protein} column contains UniProt IDs in the format \code{database|acce
 
 \seealso{
 Amino acid compositions of non-human proteins are stored under \code{extdata/aa} in directories \code{archaea}, \code{bacteria}, \code{cow}, \code{dog}, \code{mouse}, \code{rat}, and \code{yeast}.
-These files can be loaded in \code{\link{protcomp}} via the \code{aa_file} argument.
+These files can be loaded in \code{\link{human.aa}} via the \code{aa_file} argument.
 }
 
 \examples{

man/human.aa.Rd

@@ -0,0 +1,37 @@
+\encoding{UTF-8}
+\name{human.aa}
+\alias{human.aa}
+\title{Get Amino Acid Compositions of Human Proteins}
+\description{
+  Get amino acid compositions of human proteins from their UniProt IDs.
+}
+
+\usage{
+  human.aa(uniprot = NULL, aa_file = NULL,
+    stop.if.missing = FALSE, warn.if.duplicated = FALSE)
+}
+
+\arguments{
+  \item{uniprot}{character, UniProt IDs of proteins}
+  \item{aa_file}{character, file name}
+  \item{stop.if.missing}{logical, stop with an error if there are UniProt IDs that can't be found?}
+  \item{warn.if.duplicated}{logical, emit a warning if duplicate UniProt IDs are detected?}
+}
+
+\details{
+This function retrieves the amino acid compositions of one or more proteins specified by \code{uniprot}.
+
+This function depends on the amino acid compositions of human proteins, which are stored in the \code{\link{canprot}} environment when the package is attached.
+If \code{aa_file} is specified, additional amino acid compositions are read from this file.
+This file should be in the same format as \code{\link{human_extra}.csv} in the installation directory of the package.
+}
+
+\value{
+The function returns a data frame with amino acid compositions of proteins.
+}
+
+\examples{
+human.aa("P24298")
+}
+
+\concept{Amino acid composition}

vignettes/introduction.Rmd

@@ -72,11 +72,11 @@ The vertical line denotes the proposed timing of the Great Oxidation Event (GOE)
 ## Human proteins in **canprot**
 
 **canprot** has a database of amino acid compositions of human proteins assembled from UniProt.
-Use `protcomp()` to get the amino acid composition.
+Use `human.aa()` to get the amino acid composition.
 This example is for alanine aminotransferase, which has a UniProt ID of [P24298](https://www.uniprot.org/uniprotkb/P24298):
-```{r protcomp_ALAT1}
-(pc <- protcomp("P24298"))
-Zc(pc$aa)
+```{r human.aa_ALAT1}
+(aa <- human.aa("P24298"))
+Zc(aa)
 ```
 
 Do you have a list of UniProt IDs for a differential expression dataset?
@@ -107,8 +107,8 @@ down <- c("J3KN67", "Q9Y490", "J3KNQ4", "E7EVA0", "Q01082", "J3KQ32",
 With those UniProt IDs for human proteins we can retrieve the amino acid compositions, then calculate a couple of chemical metrics and make some boxplots comparing the groups of differentially regulated proteins.
 
 ```{r DKM20_plot, out.width = "75%", fig.align = "center", fig.width = 7, fig.height = 5}
-aa_down <- protcomp(down)$aa
-aa_up <- protcomp(up)$aa
+aa_down <- human.aa(down)
+aa_up <- human.aa(up)
 bp_names <- paste0(c("Down (", "Up ("), c(nrow(aa_down), nrow(aa_up)), c(")", ")"))
 
 par(mfrow = c(1, 2))