diff --git a/.Rbuildignore b/.Rbuildignore
new file mode 100644
index 0000000..85d152d
--- /dev/null
+++ b/.Rbuildignore
@@ -0,0 +1,6 @@
+^.*\.Rproj$
+^\.Rproj\.user$
+^\.git$
+^\.gitignore$
+^README\.md$
+^\.github$ 
\ No newline at end of file
diff --git a/DESCRIPTION b/DESCRIPTION
new file mode 100644
index 0000000..1eb8bc7
--- /dev/null
+++ b/DESCRIPTION
@@ -0,0 +1,16 @@
+Package: dsCodaClient
+Title: DataSHIELD client-side functions for compositional data analysis
+Description: DataSHIELD client functions for compositional data analysis using the compositions package.
+Version: 0.1.0
+Author: Xavier Escriba Montagut and Stuart Wheater
+Maintainer: Xavier Escriba Montagut and Stuart Wheater <stuart.wheater@arjuna.com>
+License: GPL-3
+Depends:
+    R (>= 3.5.0),
+    DSI (>= 1.1.0)
+Suggests:
+    knitr,
+    testthat
+RoxygenNote: 7.3.2
+Encoding: UTF-8
+VignetteBuilder: knitr 
diff --git a/NAMESPACE b/NAMESPACE
new file mode 100644
index 0000000..ee02a4f
--- /dev/null
+++ b/NAMESPACE
@@ -0,0 +1,4 @@
+# Generated by roxygen2: do not edit by hand
+
+export(ds.acomp)
+export(ds.lrEM)
diff --git a/R/ds.acomp.R b/R/ds.acomp.R
new file mode 100644
index 0000000..ef72cec
--- /dev/null
+++ b/R/ds.acomp.R
@@ -0,0 +1,117 @@
+#' @title Creates a server-side Aitchison composition object for compositional data analysis
+#' @description Creates a server-side Aitchison composition object using the compositions::acomp() function
+#' @details This function creates a server-side Aitchison composition object that can be used
+#' for compositional data analysis. It wraps the compositions::acomp() function for use in DataSHIELD.
+#' 
+#' Server function called: \code{acompDS}. 
+#' 
+#' @param X character string specifying the name of the server-side dataset or vector 
+#' containing the compositional data.
+#' @param parts vector containing the indices or names of the columns to be used.
+#' If NULL (default), all columns will be used.
+#' @param total numeric value specifying the total amount to be used, typically 1 or 100.
+#' Default is 1.
+#' @param warn.na logical value indicating whether the user should be warned in case of 
+#' NA, NaN or 0 coding different types of missing values. Default is FALSE.
+#' @param detectionlimit a number, vector or matrix of positive numbers giving the 
+#' detection limit of all values, all columns or each value, respectively. Default is NULL.
+#' @param BDL the code for 'Below Detection Limit' in X. Default is NULL.
+#' @param MAR the code for 'Missing At Random' in X. Default is NULL.
+#' @param MNAR the code for 'Missing Not At Random' in X. Default is NULL.
+#' @param SZ the code for 'Structural Zero' in X. Default is NULL.
+#' @param objectname character string specifying the name of the new server-side object 
+#' which will store the Aitchison composition object.
+#' @param datasources a list of \code{\link{DSConnection-class}} objects obtained after login. 
+#' If the \code{datasources} argument is not specified, the default set of connections will be used: 
+#' see \code{\link{datashield.connections_default}}.
+#' @return \code{ds.acomp} returns a server-side Aitchison composition object for use in
+#' compositional data analysis.
+#' @author DataSHIELD Development Team
+#' @examples
+#' \dontrun{
+#'
+#'   # connecting to the DataSHIELD servers
+#' 
+#'   require('DSI')
+#'   require('DSOpal')
+#'   require('dsBaseClient')
+#'   require('dsCodaClient')
+#'
+#'   builder <- DSI::newDSLoginBuilder()
+#'   builder$append(server = "study1", 
+#'                  url = "http://192.168.56.100:8080/", 
+#'                  user = "administrator", password = "datashield_test&", 
+#'                  table = "COMPOSITIONAL.DATA1", driver = "OpalDriver")
+#'   builder$append(server = "study2", 
+#'                  url = "http://192.168.56.100:8080/", 
+#'                  user = "administrator", password = "datashield_test&", 
+#'                  table = "COMPOSITIONAL.DATA2", driver = "OpalDriver")
+#'   logindata <- builder$build()
+#'   
+#'   connections <- DSI::datashield.login(logins = logindata, assign = TRUE, symbol = "D") 
+#'   
+#'   # Create a server-side Aitchison composition object
+#'   # Using all columns of the dataset
+#'   dsCodaClient::ds.acomp(X = 'D', 
+#'                          objectname = 'comp_object',
+#'                          datasources = connections)
+#'
+#'   # Using specific columns
+#'   dsCodaClient::ds.acomp(X = 'D', 
+#'                          parts = c(1, 2, 3),
+#'                          total = 100,
+#'                          objectname = 'comp_object_subset',
+#'                          datasources = connections)
+#'
+#'   # Using column names
+#'   dsCodaClient::ds.acomp(X = 'D', 
+#'                          parts = c('component1', 'component2', 'component3'),
+#'                          objectname = 'comp_object_named',
+#'                          datasources = connections)
+#'   
+#'   # clear the DataSHIELD R sessions and logout
+#'   datashield.logout(connections)
+#' }
+#'
+#' @export
+ds.acomp <- function(
+  X = NULL,
+  parts = NULL,
+  total = 1,
+  warn.na = FALSE,
+  detectionlimit = NULL,
+  BDL = NULL,
+  MAR = NULL,
+  MNAR = NULL,
+  SZ = NULL,
+  objectname = NULL,
+  datasources = NULL
+) {
+
+  # look for DS connections
+  # if one not provided then get current
+  if (is.null(datasources)) {
+    datasources <- DSI::datashield.connections_find()
+  }
+
+  # verify that 'X' was set
+  if (is.null(X)) {
+    stop("Please provide a valid compositional dataset parameter (X)", call. = FALSE)
+  }
+
+  # handle objectname - use default if not provided
+  if (is.null(objectname)) {
+    objectname <- "acomp_object"
+    warning("No objectname provided, using default: 'acomp_object'", call. = FALSE)
+  }
+
+  # call the server side function
+  calltext <- call("acompDS", X, parts, total, warn.na, detectionlimit, BDL, MAR, MNAR, SZ)
+
+  # call assign function to create the object on the server
+  output <- DSI::datashield.assign(conns = datasources, value = calltext, symbol = objectname)
+
+  # return the output
+  return(output)
+
+}
\ No newline at end of file
diff --git a/R/ds.lrEM.R b/R/ds.lrEM.R
new file mode 100644
index 0000000..04647e6
--- /dev/null
+++ b/R/ds.lrEM.R
@@ -0,0 +1,147 @@
+#' @title Creates a server-side log-ratio EM algorithm object for compositional data imputation
+#' @description Creates a server-side object using the zCompositions::lrEM() function
+#' @details This function creates a server-side log-ratio EM algorithm object that can be used
+#' for imputing left-censored data (e.g. values below detection limit, rounded zeros) in 
+#' compositional data sets. It wraps the zCompositions::lrEM() function for use in DataSHIELD.
+#' 
+#' Server function called: \code{lrEMDS}. 
+#' 
+#' @param X character string specifying the name of the server-side dataset or matrix 
+#' containing the compositional data.
+#' @param label unique label (numeric or character) used to denote unobserved values in X. 
+#' Default is NULL.
+#' @param dl numeric vector or matrix of detection limits/thresholds. These must be given on 
+#' the same scale as X. Default is NULL.
+#' @param rob logical value. FALSE provides maximum-likelihood estimates of model parameters, 
+#' TRUE provides robust parameter estimates. Default is FALSE.
+#' @param ini.cov initial estimation of either the log-ratio covariance matrix (ML estimation) 
+#' or unobserved data (robust estimation). It can be based on either complete observations 
+#' ("complete.obs") or multiplicative simple replacement ("multRepl"). Default is "complete.obs".
+#' @param frac if ini.cov="multRepl", fraction parameter for initial multiplicative simple 
+#' replacement of left-censored data expressed in proportions. Default is 0.65.
+#' @param tolerance convergence criterion for the EM algorithm. Default is 0.0001.
+#' @param max.iter maximum number of iterations for the EM algorithm. Default is 50.
+#' @param rlm.maxit if rob=TRUE, maximum number of iterations for the embedded robust 
+#' regression estimation. Default is 150.
+#' @param imp.missing if TRUE then unobserved data identified by label are treated as 
+#' missing data. Default is FALSE.
+#' @param suppress.print suppress printed feedback. Default is FALSE.
+#' @param closure closure value used to add a residual part if needed when ini.cov="multRepl" 
+#' is used. Default is NULL.
+#' @param z.warning threshold for the proportion of modified zeros per part above which 
+#' a warning is issued. Default is 0.8.
+#' @param z.delete logical value indicating whether rows containing only zeros should be 
+#' deleted prior to imputation. Default is TRUE.
+#' @param objectname character string specifying the name of the new server-side object 
+#' which will store the log-ratio EM imputation result.
+#' @param datasources a list of \code{\link{DSConnection-class}} objects obtained after login. 
+#' If the \code{datasources} argument is not specified, the default set of connections will be used: 
+#' see \code{\link{datashield.connections_default}}.
+#' @return \code{ds.lrEM} returns a server-side object containing the imputed compositional 
+#' data set for use in compositional data analysis.
+#' @author DataSHIELD Development Team
+#' @examples
+#' \dontrun{
+#'
+#'   # connecting to the DataSHIELD servers
+#' 
+#'   require('DSI')
+#'   require('DSOpal')
+#'   require('dsBaseClient')
+#'   require('dsCodaClient')
+#'
+#'   builder <- DSI::newDSLoginBuilder()
+#'   builder$append(server = "study1", 
+#'                  url = "http://192.168.56.100:8080/", 
+#'                  user = "administrator", password = "datashield_test&", 
+#'                  table = "COMPOSITIONAL.DATA1", driver = "OpalDriver")
+#'   builder$append(server = "study2", 
+#'                  url = "http://192.168.56.100:8080/", 
+#'                  user = "administrator", password = "datashield_test&", 
+#'                  table = "COMPOSITIONAL.DATA2", driver = "OpalDriver")
+#'   logindata <- builder$build()
+#'   
+#'   connections <- DSI::datashield.login(logins = logindata, assign = TRUE, symbol = "D") 
+#'   
+#'   # Create a server-side log-ratio EM imputation
+#'   # Using default parameters
+#'   dsCodaClient::ds.lrEM(X = 'D', 
+#'                         label = 0,
+#'                         objectname = 'imputed_data',
+#'                         datasources = connections)
+#'
+#'   # Using specific detection limits
+#'   dsCodaClient::ds.lrEM(X = 'D', 
+#'                         label = 0,
+#'                         dl = c(0.1, 0.1, 0.1),
+#'                         rob = FALSE,
+#'                         tolerance = 0.001,
+#'                         objectname = 'imputed_data_custom',
+#'                         datasources = connections)
+#'
+#'   # Using robust estimation
+#'   dsCodaClient::ds.lrEM(X = 'D', 
+#'                         label = 0,
+#'                         dl = c(0.1, 0.1, 0.1),
+#'                         rob = TRUE,
+#'                         ini.cov = "multRepl",
+#'                         frac = 0.5,
+#'                         objectname = 'robust_imputed_data',
+#'                         datasources = connections)
+#'   
+#'   # clear the DataSHIELD R sessions and logout
+#'   datashield.logout(connections)
+#' }
+#'
+#' @export
+ds.lrEM <- function(
+  X = NULL,
+  label = NULL,
+  dl = NULL,
+  rob = FALSE,
+  ini.cov = c("complete.obs", "multRepl"),
+  frac = 0.65,
+  tolerance = 0.0001,
+  max.iter = 50,
+  rlm.maxit = 150,
+  imp.missing = FALSE,
+  suppress.print = FALSE,
+  closure = NULL,
+  z.warning = 0.8,
+  z.delete = TRUE,
+  objectname = NULL,
+  datasources = NULL
+) {
+
+  # look for DS connections
+  # if one not provided then get current
+  if (is.null(datasources)) {
+    datasources <- DSI::datashield.connections_find()
+  }
+
+  # verify that 'X' was set
+  if (is.null(X)) {
+    stop("Please provide a valid compositional dataset parameter (X)", call. = FALSE)
+  }
+
+  # handle objectname - use default if not provided
+  if (is.null(objectname)) {
+    objectname <- "lrEM_object"
+    warning("No objectname provided, using default: 'lrEM_object'", call. = FALSE)
+  }
+
+  # match ini.cov argument
+  ini.cov <- match.arg(ini.cov)
+
+  # call the server side function
+  calltext <- call("lrEMDS", X, label, dl, rob, ini.cov, frac, tolerance, 
+                   max.iter, rlm.maxit, imp.missing, suppress.print, closure,
+                   z.warning, z.delete)
+
+  # call assign function to create the object on the server
+  output <- DSI::datashield.assign(conns = datasources, value = calltext, symbol = objectname)
+
+  # return the output
+  return(output)
+
+} 
\ No newline at end of file
diff --git a/man/ds.acomp.Rd b/man/ds.acomp.Rd
new file mode 100644
index 0000000..a4f37e0
--- /dev/null
+++ b/man/ds.acomp.Rd
@@ -0,0 +1,114 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/ds.acomp.R
+\name{ds.acomp}
+\alias{ds.acomp}
+\title{Creates a server-side Aitchison composition object for compositional data analysis}
+\usage{
+ds.acomp(
+  X = NULL,
+  parts = NULL,
+  total = 1,
+  warn.na = FALSE,
+  detectionlimit = NULL,
+  BDL = NULL,
+  MAR = NULL,
+  MNAR = NULL,
+  SZ = NULL,
+  objectname = NULL,
+  datasources = NULL
+)
+}
+\arguments{
+\item{X}{character string specifying the name of the server-side dataset or vector 
+containing the compositional data.}
+
+\item{parts}{vector containing the indices or names of the columns to be used.
+If NULL (default), all columns will be used.}
+
+\item{total}{numeric value specifying the total amount to be used, typically 1 or 100.
+Default is 1.}
+
+\item{warn.na}{logical value indicating whether the user should be warned in case of 
+NA, NaN or 0 coding different types of missing values. Default is FALSE.}
+
+\item{detectionlimit}{a number, vector or matrix of positive numbers giving the 
+detection limit of all values, all columns or each value, respectively. Default is NULL.}
+
+\item{BDL}{the code for 'Below Detection Limit' in X. Default is NULL.}
+
+\item{MAR}{the code for 'Missing At Random' in X. Default is NULL.}
+
+\item{MNAR}{the code for 'Missing Not At Random' in X. Default is NULL.}
+
+\item{SZ}{the code for 'Structural Zero' in X. Default is NULL.}
+
+\item{objectname}{character string specifying the name of the new server-side object 
+which will store the Aitchison composition object.}
+
+\item{datasources}{a list of \code{\link{DSConnection-class}} objects obtained after login. 
+If the \code{datasources} argument is not specified, the default set of connections will be used: 
+see \code{\link{datashield.connections_default}}.}
+}
+\value{
+\code{ds.acomp} returns a server-side Aitchison composition object for use in
+compositional data analysis.
+}
+\description{
+Creates a server-side Aitchison composition object using the compositions::acomp() function
+}
+\details{
+This function creates a server-side Aitchison composition object that can be used
+for compositional data analysis. It wraps the compositions::acomp() function for use in DataSHIELD.
+
+Server function called: \code{acompDS}.
+}
+\examples{
+\dontrun{
+
+  # connecting to the DataSHIELD servers
+
+  require('DSI')
+  require('DSOpal')
+  require('dsBaseClient')
+  require('dsCodaClient')
+
+  builder <- DSI::newDSLoginBuilder()
+  builder$append(server = "study1", 
+                 url = "http://192.168.56.100:8080/", 
+                 user = "administrator", password = "datashield_test&", 
+                 table = "COMPOSITIONAL.DATA1", driver = "OpalDriver")
+  builder$append(server = "study2", 
+                 url = "http://192.168.56.100:8080/", 
+                 user = "administrator", password = "datashield_test&", 
+                 table = "COMPOSITIONAL.DATA2", driver = "OpalDriver")
+  logindata <- builder$build()
+  
+  connections <- DSI::datashield.login(logins = logindata, assign = TRUE, symbol = "D") 
+  
+  # Create a server-side Aitchison composition object
+  # Using all columns of the dataset
+  dsCodaClient::ds.acomp(X = 'D', 
+                         objectname = 'comp_object',
+                         datasources = connections)
+
+  # Using specific columns
+  dsCodaClient::ds.acomp(X = 'D', 
+                         parts = c(1, 2, 3),
+                         total = 100,
+                         objectname = 'comp_object_subset',
+                         datasources = connections)
+
+  # Using column names
+  dsCodaClient::ds.acomp(X = 'D', 
+                         parts = c('component1', 'component2', 'component3'),
+                         objectname = 'comp_object_named',
+                         datasources = connections)
+  
+  # clear the DataSHIELD R sessions and logout
+  datashield.logout(connections)
+}
+
+}
+\author{
+DataSHIELD Development Team
+}
diff --git a/man/ds.lrEM.Rd b/man/ds.lrEM.Rd
new file mode 100644
index 0000000..dce6830
--- /dev/null
+++ b/man/ds.lrEM.Rd
@@ -0,0 +1,144 @@
+% Generated by roxygen2: do not edit by hand
+% Please edit documentation in R/ds.lrEM.R
+\name{ds.lrEM}
+\alias{ds.lrEM}
+\title{Creates a server-side log-ratio EM algorithm object for compositional data imputation}
+\usage{
+ds.lrEM(
+  X = NULL,
+  label = NULL,
+  dl = NULL,
+  rob = FALSE,
+  ini.cov = c("complete.obs", "multRepl"),
+  frac = 0.65,
+  tolerance = 1e-04,
+  max.iter = 50,
+  rlm.maxit = 150,
+  imp.missing = FALSE,
+  suppress.print = FALSE,
+  closure = NULL,
+  z.warning = 0.8,
+  z.delete = TRUE,
+  objectname = NULL,
+  datasources = NULL
+)
+}
+\arguments{
+\item{X}{character string specifying the name of the server-side dataset or matrix 
+containing the compositional data.}
+
+\item{label}{unique label (numeric or character) used to denote unobserved values in X. 
+Default is NULL.}
+
+\item{dl}{numeric vector or matrix of detection limits/thresholds. These must be given on 
+the same scale as X. Default is NULL.}
+
+\item{rob}{logical value. FALSE provides maximum-likelihood estimates of model parameters, 
+TRUE provides robust parameter estimates. Default is FALSE.}
+
+\item{ini.cov}{initial estimation of either the log-ratio covariance matrix (ML estimation) 
+or unobserved data (robust estimation). It can be based on either complete observations 
+("complete.obs") or multiplicative simple replacement ("multRepl"). Default is "complete.obs".}
+
+\item{frac}{if ini.cov="multRepl", fraction parameter for initial multiplicative simple 
+replacement of left-censored data expressed in proportions. Default is 0.65.}
+
+\item{tolerance}{convergence criterion for the EM algorithm. Default is 0.0001.}
+
+\item{max.iter}{maximum number of iterations for the EM algorithm. Default is 50.}
+
+\item{rlm.maxit}{if rob=TRUE, maximum number of iterations for the embedded robust 
+regression estimation. Default is 150.}
+
+\item{imp.missing}{if TRUE then unobserved data identified by label are treated as 
+missing data. Default is FALSE.}
+
+\item{suppress.print}{suppress printed feedback. Default is FALSE.}
+
+\item{closure}{closure value used to add a residual part if needed when ini.cov="multRepl" 
+is used. Default is NULL.}
+
+\item{z.warning}{threshold for the proportion of modified zeros per part above which 
+a warning is issued. Default is 0.8.}
+
+\item{z.delete}{logical value indicating whether rows containing only zeros should be 
+deleted prior to imputation. Default is TRUE.}
+
+\item{objectname}{character string specifying the name of the new server-side object 
+which will store the log-ratio EM imputation result.}
+
+\item{datasources}{a list of \code{\link{DSConnection-class}} objects obtained after login. 
+If the \code{datasources} argument is not specified, the default set of connections will be used: 
+see \code{\link{datashield.connections_default}}.}
+}
+\value{
+\code{ds.lrEM} returns a server-side object containing the imputed compositional 
+data set for use in compositional data analysis.
+}
+\description{
+Creates a server-side object using the zCompositions::lrEM() function
+}
+\details{
+This function creates a server-side log-ratio EM algorithm object that can be used
+for imputing left-censored data (e.g. values below detection limit, rounded zeros) in 
+compositional data sets. It wraps the zCompositions::lrEM() function for use in DataSHIELD.
+
+Server function called: \code{lrEMDS}.
+}
+\examples{
+\dontrun{
+
+  # connecting to the DataSHIELD servers
+
+  require('DSI')
+  require('DSOpal')
+  require('dsBaseClient')
+  require('dsCodaClient')
+
+  builder <- DSI::newDSLoginBuilder()
+  builder$append(server = "study1", 
+                 url = "http://192.168.56.100:8080/", 
+                 user = "administrator", password = "datashield_test&", 
+                 table = "COMPOSITIONAL.DATA1", driver = "OpalDriver")
+  builder$append(server = "study2", 
+                 url = "http://192.168.56.100:8080/", 
+                 user = "administrator", password = "datashield_test&", 
+                 table = "COMPOSITIONAL.DATA2", driver = "OpalDriver")
+  logindata <- builder$build()
+  
+  connections <- DSI::datashield.login(logins = logindata, assign = TRUE, symbol = "D") 
+  
+  # Create a server-side log-ratio EM imputation
+  # Using default parameters
+  dsCodaClient::ds.lrEM(X = 'D', 
+                        label = 0,
+                        objectname = 'imputed_data',
+                        datasources = connections)
+
+  # Using specific detection limits
+  dsCodaClient::ds.lrEM(X = 'D', 
+                        label = 0,
+                        dl = c(0.1, 0.1, 0.1),
+                        rob = FALSE,
+                        tolerance = 0.001,
+                        objectname = 'imputed_data_custom',
+                        datasources = connections)
+
+  # Using robust estimation
+  dsCodaClient::ds.lrEM(X = 'D', 
+                        label = 0,
+                        dl = c(0.1, 0.1, 0.1),
+                        rob = TRUE,
+                        ini.cov = "multRepl",
+                        frac = 0.5,
+                        objectname = 'robust_imputed_data',
+                        datasources = connections)
+  
+  # clear the DataSHIELD R sessions and logout
+  datashield.logout(connections)
+}
+
+}
+\author{
+DataSHIELD Development Team
+}