R/distances.R

Defines functions norm.connect norm.distances norm is.distances dist.point.default distances.pdb distances.connect distances.atoms distances.coords distances.default dist.point distances

Documented in distances distances.atoms distances.connect distances.coords distances.default distances.pdb dist.point dist.point.default is.distances norm norm.connect norm.distances

#' Inter-Atomic Distances
#' 
#' Computes inter-atomic distance vectors.
#' 
#' The purpose of the \sQuote{distances} class is to store the inter-atomic 
#' distance vectors and facilitate their manipulation when passing from the 
#' Cartesian to fractional references and vice versa.\cr
#' The default method of the \code{distances} function is actually a builder
#'   allowing to create a \sQuote{distances} object from its different components, i.e.:
#'   \code{dx1}, \code{dx2}, \code{dx3}, and \code{basis}.
#'   All the arguments have to be specified except 'basis' which
#'   by default is set to "xyz" (Cartesian reference).
#' 
#' For objects of class \sQuote{coords}, \sQuote{atoms}, \sQuote{pdb},
#' two sets of atomic coordinates, defined by \code{sel1} and \code{sel2},
#' are extracted and inter-atomic distance vectors are computed 
#' between these two sets.
#' For objects of class \sQuote{connect}, distances are computed for the atoms
#'   specified by the \sQuote{connect} object.
#' 
#' The method for the \sQuote{dist.point} function computes the inter-atomic distances
#' between the atoms and a specified point.
#' 
#' The method of the \code{norm} function for objects of class \sQuote{distances}
#'   computes the norm of the distances vectors.
#' \code{type} specify how to project the distance vectors before 
#'   computing the norms. By default no projection is perform.
#'   The three dx, dy, and dz components of the distance vectors
#'   are used to compute the norm. 
#' \code{type} can take the following values:
#'    \itemize{
#'    \item   x: The distance vectors are projected over x.
#'    \item   y: The distance vectors are projected over y.
#'    \item   z: The distance vectors are projected over z.
#'    \item  xy: The distance vectors are projected in the xy-plan.
#'    \item  yz: The distance vectors are projected in the yz-plan.
#'    \item  zx: The distance vectors are projected in the zx-plan.
#'    \item xyz: The distance vectors are not projected 
#'      (The three components of the distance vectors are used to compute the norm).}
#' \code{is.distances} tests if x is an object of class \sQuote{distances};
#' the test is limited to the class attribute.
#' 
#' @return The \code{distances} and \code{dist.point} functions return an object
#'   of class \sQuote{distances} containing inter-atomic distance vectors.
#' The \code{norm} function return an array, with the same dimensions as
#'   the \code{dx1}, \code{dx2}, \code{dx3} components of the \sQuote{distances} object
#'   for which the norms have been computed, containing the norm of the distance vectors. 
#' \cr\cr
#' \code{is.distances} returns TRUE if x is an object of class \sQuote{distances}
#' and FALSE otherwise.
#' 
#' @param dx1,dx2,dx3 numeric arrays containing the first, second and third components of the distance vectors.
#' @param basis a single element character vector indicating the type of basis vector used to express the coordinates.
#' @param x,y,z an R object containing atomic coordinates, or a a connect object (for the corresponding method);
#' @param sel1,sel2 integer or logical vectors defining two atomic selections between which the distance vectors are computed.
#' @param type a single element character vector indicating how to project the distances vectors before computing the norms. See details.
#' @param data an object of type pdb, atoms or coords, containing atomic coordinates.
#' @param subset enables sub-setting the coords object;
#' @param \dots further arguments passed to or from other methods.
#' 
#' @seealso 
#' \code{\link{coords}}, \code{\link{basis}}, \code{\link{xyz2abc}}, \code{\link{abc2xyz}}
#' 
#' @examples 
#' x <- read.pdb(system.file("examples/PCBM_ODCB.pdb", package="Rpdb"))
#' is.DCB7 <- x$atoms$resname == "DCB" & x$atoms$resid == 7
#' is.DCB8 <- x$atoms$resname == "DCB" & x$atoms$resid == 8
#' d <- distances(x, is.DCB7, is.DCB8)
#' norm(d, type = "xyz")
#' norm(d, type = "xy")
#' norm(d, type = "x")
#
#' d <- dist.point(x, c(0,0,0), subset = is.DCB7)
#' norm(d, type = "xyz")
#' 
#' @keywords classes manip
#' 
#' @name distances
#' @export
distances <- function(...)
  UseMethod("distances")

#' @rdname distances
#' @export
dist.point <- function(...)
  UseMethod("dist.point")

#' @rdname distances
#' @export
distances.default <- function(dx1 = numeric(0), dx2 = numeric(0), dx3 = numeric(0), basis = "xyz", ...) {
  if( ! is.numeric(dx1) || ! is.numeric(dx2) || ! is.numeric(dx3))
    stop("'dx1', 'dx2' and 'dx3' must be numeric");
  if(is.null(dim(dx1))) {
    if( ! is.null(dim(dx2)) && ! is.null(dim(dx3)))
      stop("'dx1', 'dx2' and 'dx3' must have the same length/dimensions");
    if(length(dx1) != length(dx2) || length(dx1) != length(dx3))
      stop("'dx1', 'dx2' and 'dx3' must have the same length");
  }
  else {
    if(any(dim(dx1) != dim(dx2)) || any(dim(dx1) != dim(dx3)))
      stop("'dx1', 'dx2' and 'dx3' must have the same dimensions");
  }
  check.basis(basis);
  
  to.return <- list(dx1 = dx1, dx2 = dx2, dx3 = dx3);
  class(to.return) <- c("distances", "list");
  attr(to.return, "basis") <- basis;
  return(to.return)
}

#' @rdname distances
#' @export
distances.coords <- function(x, sel1, sel2, ...){
  if(missing(sel1) | missing(sel2))
    stop("Please specify 'sel1' and 'sel2'")
  if(is.numeric(sel1)) {
    check.idOrLogical(sel1, natom(x), "'sel1'");
  }
  if(is.numeric(sel2)) {
    check.idOrLogical(sel2, natom(x), "'sel2'");
  }
  if(is.logical(sel1) & length(sel1) != natom(x))
    stop("'sel1' length must be equal to natom(x)")
  if(is.logical(sel2) & length(sel2) != natom(x))
    stop("'sel2' length must be equal to natom(x)")  

  xyz1 <- x[sel1,]
  xyz2 <- x[sel2,]
  
  dx1 <- t(outer(xyz2$x1, xyz1$x1, "-"))
  dx2 <- t(outer(xyz2$x2, xyz1$x2, "-"))
  dx3 <- t(outer(xyz2$x3, xyz1$x3, "-"))
  
  dimnames(dx1) <- list(sel1 = NULL, sel2 = NULL)
  dimnames(dx2) <- list(sel1 = NULL, sel2 = NULL)
  dimnames(dx3) <- list(sel1 = NULL, sel2 = NULL)
  
  to.return <- distances.default(dx1, dx2, dx3, basis = basis(x))
  return(to.return)
}

#' @rdname distances
#' @export
distances.atoms <- function(x, sel1, sel2, ...)
  distances.coords(coords(x), sel1 = sel1, sel2 = sel2, ...)

#' @rdname distances
#' @export
distances.connect = function(x, data, ...) {
	sel1 = x$eleid.1;
	sel2 = x$eleid.2;
	coords = coords(data);
	d = coords[sel1,] - coords[sel2,];
	d = data.frame(dx1 = d[,1], dx2 = d[,2], dx3 = d[,3]);
	attr(d, "basis")   = basis(data);
	class(d) = c("distances", "data.frame");
	return(d);
}

#' @rdname distances
#' @export
distances.pdb <- function(x, sel1, sel2, ...)
  distances.atoms(x$atoms, sel1 = sel1, sel2 = sel2, ...)

### Distance to a Point
#' @rdname distances
#' @export
dist.point.default <- function(data, x, y = NULL, z = NULL, subset = NULL, ...) {
	if(is.null(y)) {
		if(length(x) != 3) stop("Missing coordinates for given point!");
		y = x[2]; z = x[3]; x = x[1];
	}
	xyz = coords(data);
	if(is.null(xyz)) {
		stop("Data must be an object that contains coordinates, e.g.",
			"pdb, atoms or coords!");
	}
	basis = basis(xyz);
	if( ! is.null(subset)) {
		xyz = subset(xyz, subset);
	}
	
	dx1 = xyz$x1 - x;
	dx2 = xyz$x2 - y;
	dx3 = xyz$x3 - z;
	
	to.return <- distances.default(dx1, dx2, dx3, basis=basis);
	return(to.return)
}

#' @rdname distances
#' @export
is.distances <- function(x)
	inherits(x, "distances");

#' @rdname distances
#' @export
norm <- function(...)
  UseMethod("norm")

#' @rdname distances
#' @export
norm.distances <- function(x, type = "xyz", ...){
  if(basis(x) == "abc")
    stop("Please provide Cartesian coordinates. See 'abc2xyz'.")
  
  to.return <- switch(type,
    x   = x$dx1,
    y   = x$dx2,
    z   = x$dx3,
    xy  = sqrt(x$dx1^2 + x$dx2^2),
    yz  = sqrt(x$dx2^2 + x$dx3^2),
    zx  = sqrt(x$dx3^2 + x$dx1^2),
    xyz = sqrt(x$dx1^2 + x$dx2^2 + x$dx3^2))

  return(to.return)
}

#' @rdname distances
#' @export
norm.connect = function(x, data, type = "xyz", ...) {
	d = distances(x, data = data);
	norm.distances(d, type = type, ...);
}

Try the Rpdb package in your browser

Any scripts or data that you put into this service are public.

Rpdb documentation built on May 12, 2026, 5:06 p.m.