Finalizing MAGNET v1.2.0 file upload

Finalizing MAGNET v1.2.0 file upload

bin/MAGNET-1.2.0/RAxMLRunChecker

@@ -0,0 +1,341 @@
+#!/bin/sh
+
+##########################################################################################
+#  __  o  __   __   __  |__   __                                                         #
+# |__) | |  ' (__( |  ) |  ) (__(                                                        # 
+# |                                                                                      #
+#                                                                                        #
+# File: RAxMLRunChecker.sh                                                               #
+  VERSION="v1.3.0"                                                                       #
+# Author: Justin C. Bagley                                                               #
+# Date: Created by Justin Bagley on/before November 29, 2018.                            #
+# Last update: March 14, 2019                                                            #
+# Copyright (c) 2018-2019 Justin C. Bagley. All rights reserved.                         #
+# Please report bugs to <jbagley@jsu.edu>.                                              #
+#                                                                                        #
+# Description:                                                                           #
+# SHELL SCRIPT THAT COUNTS NUMBER OF LOCI/PARTITIONS WITH COMPLETED RAxML RUNS DURING    #
+# OR AFTER A RUN OF THE MAGNET PIPELINE, AND COLLATES RUN INFORMATION                    #
+#                                                                                        #
+##########################################################################################
+
+# Provide a variable with the location of this script.
+SCRIPT_PATH="$( cd "$( dirname "${BASH_SOURCE[0]}" )" && pwd )"
+
+# Source Scripting Utilities
+# -----------------------------------
+# These shared utilities provide many functions which are needed to provide
+# the functionality in this boilerplate. This script will fail if they can
+# not be found.
+# -----------------------------------
+
+UTILS_LOCATION="${SCRIPT_PATH}/../lib/utils.sh" # Update this path to find the utilities.
+
+if [[ -f "${UTILS_LOCATION}" ]]; then
+  source "${UTILS_LOCATION}"
+else
+  echo "Please find the file util.sh and add a reference to it in this script. Exiting..."
+  exit 1
+fi
+
+
+# Source shared functions and variables
+# -----------------------------------
+
+FUNCS_LOCATION="${SCRIPT_PATH}/../lib/sharedFunctions.sh" # Update this path to find the shared functions.
+VARS_LOCATION="${SCRIPT_PATH}/../lib/sharedVariables.sh" # Update this path to find the shared variables.
+
+if [[ -f "${FUNCS_LOCATION}" ]] && [[ -f "${VARS_LOCATION}" ]]; then
+  source "${FUNCS_LOCATION}" ;
+  source "${VARS_LOCATION}" ;
+else
+  echo "Please find the files sharedFunctions.sh and sharedVariables.sh and add references to them in this script. Exiting... "
+  exit 1
+fi
+
+
+# trapCleanup Function
+# -----------------------------------
+# Any actions that should be taken if the script is prematurely
+# exited.  Always call this function at the top of your script.
+# -----------------------------------
+function trapCleanup() {
+  echo ""
+  # Delete temp files, if any
+  if is_dir "${tmpDir}"; then
+    rm -r "${tmpDir}"
+  fi
+  die "Exit trapped. In function: '${FUNCNAME[*]}'"
+}
+
+# safeExit
+# -----------------------------------
+# Non destructive exit for when script exits naturally.
+# Usage: Add this function at the end of every script.
+# -----------------------------------
+function safeExit() {
+  # Delete temp files, if any
+  if is_dir "${tmpDir}"; then
+    rm -r "${tmpDir}"
+  fi
+  trap - INT TERM EXIT
+  exit
+}
+
+# Set Flags
+# -----------------------------------
+# Flags which can be overridden by user input.
+# Default values are below
+# -----------------------------------
+quiet=false
+printLog=false
+verbose=false
+force=false
+strict=false
+debug=false
+args=()
+
+# Set Temp Directory
+# -----------------------------------
+# Create temp directory with three random numbers and the process ID
+# in the name.  This directory is removed automatically at exit.
+# -----------------------------------
+tmpDir="/tmp/${SCRIPT_NAME}.$RANDOM.$RANDOM.$RANDOM.$$"
+(umask 077 && mkdir "${tmpDir}") || {
+  die "Could not create temporary directory! Exiting."
+}
+
+# Logging
+# -----------------------------------
+# Log is only used when the '-l' flag is set.
+#
+# To never save a logfile change variable to '/dev/null'
+# Save to Desktop use: $HOME/Desktop/${SCRIPT_BASENAME}.log
+# Save to standard user log location use: $HOME/Library/Logs/${SCRIPT_BASENAME}.log
+# -----------------------------------
+logFile="$HOME/Library/Logs/${SCRIPT_BASENAME}.log"
+
+# Check for Dependencies
+# -----------------------------------
+# Arrays containing package dependencies needed to execute this script.
+# The script will fail if dependencies are not installed.  For Mac users,
+# most dependencies can be installed automatically using the package
+# manager 'Homebrew'.  Mac applications will be installed using
+# Homebrew Casks. Ruby and gems via RVM.
+# -----------------------------------
+export homebrewDependencies=()
+export caskDependencies=()
+export gemDependencies=()
+
+
+
+
+function RAxMLRunChecker () {
+
+######################################## START ###########################################
+##########################################################################################
+
+echo "INFO      | $(date) |----------------------------------------------------------------"
+echo "INFO      | $(date) | RAxMLRunChecker, v1.3.0 March 2019  (part of PIrANHA v0.4a4)   "
+echo "INFO      | $(date) | Copyright (c) 2018-2019 Justin C. Bagley. All rights reserved. "
+echo "INFO      | $(date) |----------------------------------------------------------------"
+
+######################################## START ###########################################
+echo "INFO      | $(date) | Starting RAxMLRunChecker pipeline... "
+echo "INFO      | $(date) | Step #1: Set up workspace and check machine type. "
+############ I. SET WORKING DIRECTORY AND CHECK MACHINE TYPE
+#USER_SPEC_PATH="$(printf '%q\n' "$(pwd)")";
+echoCDWorkingDir
+#echo "INFO      | $(date) |          Checking machine type... "
+checkMachineType
+#echo "INFO      | $(date) |               Found machine type ${machine}. "
+
+
+############ II. RUN RAXML RUN CHECKER
+echo "INFO      | $(date) | Step #2: Check RAxML runs in subfolders in current directory (assumed to be a MAGNET run folder). "
+
+	echo "INFO      | $(date) |          Estimating numbers (no.) of loci and RAxML runs... "
+	MY_N_LOCI_FOLD="$(ls -d ./locus*/ | wc -l | sed 's/^[\ ]*//g')";
+	MY_N_COMPLETED="$(ls ./locus*/RAxML_info.raxml_out | wc -l | sed 's/^[\ ]*//g')";
+	MY_N_REMAINING="$(calc $MY_N_LOCI_FOLD - $MY_N_COMPLETED)";
+
+	echo "INFO      | $(date) |          Total no. RAxML runs: $TAB$TAB$MY_N_LOCI_FOLD "
+	echo "INFO      | $(date) |          No. completed RAxML runs: $TAB$MY_N_COMPLETED "
+	echo "INFO      | $(date) |          No. remaining RAxML runs:    $TAB$MY_N_REMAINING "
+
+	if [[ -s ./completed_run_info.tmp ]]; then
+		rm ./completed_run_info.tmp ;
+	fi
+	if [[ -s ./completed_run_info.txt ]]; then
+		rm ./completed_run_info.txt ;
+	fi
+	if [[ -s ./remaining_run_info.tmp ]]; then
+		rm ./remaining_run_info.tmp ;
+	fi
+	if [[ -s ./remaining_run_info.txt ]]; then
+		rm ./remaining_run_info.txt ;
+	fi
+
+	echo "INFO      | $(date) |          Saving RAxML run info to file... "
+
+	count=1
+	echo "INFO      | $(date) |          ...  $count / $MY_N_LOCI_FOLD ..."
+(
+	for i in ./locus*/; do 
+		MY_LOCUS="$(echo $i | sed 's/\.\///g; s/\///g; s/\ //g')"; 
+		MY_COUNT_HUND_CHECK="$(calc $count / 100 | sed 's/^[0-9]*\.//g; s/^[0]\{1\}//g')"
+		if [[ "$MY_COUNT_HUND_CHECK" -eq "0" ]]; then
+			echo "INFO      | $(date) |          ...  $count / $MY_N_LOCI_FOLD ..."
+		fi
+		if [[ "$count" -eq "$MY_N_LOCI_FOLD" ]]; then
+			echo "INFO      | $(date) |          ...  $MY_N_LOCI_FOLD / $MY_N_LOCI_FOLD ..."
+		fi
+		cd "$i"; 
+			if [[ -s RAxML_bipartitions.raxml_out ]]; then 
+
+				MY_ALIGN_PATT="$(grep -h '^Alignment\ Patterns\:\ ' ./RAxML_info.raxml_out | sed 's/^.*\:\ //g')";
+				MY_SUBST_MODEL="$(grep -h '^Substitution\ Matrix\:\ ' ./RAxML_info.raxml_out | sed 's/^.*\:\ //g')";
+				MY_OPTIM_LIKE="$(grep -h 'Final\ ML\ Optimization\ Likelihood\:\ ' ./RAxML_info.raxml_out | sed 's/^.*\:\ //g')";
+				MY_ML_RUN_TIME="$(grep -h 'Overall\ execution\ time\ ' ./RAxML_info.raxml_out | sed 's/^Overall\ execution\ time\ [A-Za-z\ ]*\:\ //g; s/or\ .*//g')";
+
+				echo "$count$TAB$MY_LOCUS$TAB$MY_ALIGN_PATT$TAB$MY_SUBST_MODEL$TAB$MY_OPTIM_LIKE$TAB$MY_ML_RUN_TIME$TAB complete" >> ../completed_run_info.tmp ; 
+			fi
+			if [[ ! -s RAxML_bipartitions.raxml_out ]]; then 
+
+				echo "$count$TAB$MY_LOCUS$TAB$MY_ALIGN_PATT$TAB$MY_SUBST_MODEL$TAB incomplete" >> ../remaining_run_info.tmp ; 
+
+			fi
+		cd ..; 
+		echo "$((count++))" > ./count.tmp ;
+	done
+)
+
+
+	echo "No$TAB Locus$TAB No. Patterns$TAB Subst. Model$TAB Likelihood$TAB ML Run Time$TAB Status" > ./header.tmp ;
+	echo "No$TAB Locus$TAB No. Patterns$TAB Subst. Model$TAB Status" > ./rem_header.tmp ;
+	cat ./header.tmp ./completed_run_info.tmp > ./completed_run_info.txt ;
+	cat ./rem_header.tmp ./remaining_run_info.tmp > ./remaining_run_info.txt ;
+
+
+	echo "INFO      | $(date) |          Editing final RAxML run information files... "
+		if [[ "${machine}" = "Mac" ]]; then
+			sed -i '' 's/\ //g' ./completed_run_info.txt ;
+			sed -i '' 's/\ //g' ./remaining_run_info.txt ;
+		fi
+		if [[ "${machine}" = "Linux" ]]; then
+			sed -i 's/\ //g' ./completed_run_info.txt ;
+			sed -i 's/\ //g' ./remaining_run_info.txt ;
+		fi
+
+
+############ III. CLEAN UP WORKSPACE BY REMOVING TEMPORARY FILES.
+echo "INFO      | $(date) | Step #3: Clean up workspace. "
+echo "INFO      | $(date) |          Cleaning up workspace by removing temporary files generated during run... "
+
+	if [[ "$(ls -1 ./*.tmp 2>/dev/null | wc -l | sed 's/\ //g')" != "0"  ]]; then 
+		rm ./*.tmp ; 
+	fi
+
+echo "----------------------------------------------------------------------------------------------------------"
+echo "output file(s): ./completed_run_info.txt "
+echo "                ./remaining_run_info.txt "
+echo ""
+
+
+##########################################################################################
+######################################### END ############################################
+
+}
+
+
+
+############ SCRIPT OPTIONS
+## OPTION DEFAULTS ##
+# None at this time.
+
+############ CREATE USAGE & HELP TEXTS
+USAGE="Usage: $(basename "$0") [OPTION]...
+
+ ${bold}Options:${reset}
+  -h   help text (also: --help) echo this help text and exit
+  -V   version (also: --version) echo version of this script and exit
+
+ ${bold}OVERVIEW${reset}
+ THIS SCRIPT was designed to run in a current working directory where the MAGNET pipeline
+ in PIrANHA v0.4a4 (Bagley 2019) is being run, or has completed a run, to estimate maximum-
+ likelihood (ML) gene trees in RAxML v8+ (Stamatakis 2014) for a set of loci from DNA 
+ sequence data. Given such a workspace, this script counts the number of loci or data 
+ partitions (each assigned a separate RAxML run in MAGNET) with completed RAxML runs and 
+ collates run information. Output files include a summary of completed runs and a summary 
+ of ongoing runs.
+	This program runs on UNIX-like and Linux systems using commonly distributed utility 
+ software, with usage obtained by running the script with the -h flag. It has been tested
+ on macOS High Sierra (v10.13+) and Mojave but should work on many earlier versions or
+ Linux (tested on CentOS 6/7). There are no other dependencies.
+
+ ${bold}Usage examples:${reset}
+ Call the program using PIrANHA, as follows:
+
+    piranha -f RAxMLRunChecker  			    Run the software
+    piranha -f RAxMLRunChecker --args='-h'      Print this help text
+
+ ${bold}CITATION${reset}
+ Bagley, J.C. 2019. PIrANHA v0.4a4. GitHub repository, Available at:
+	<https://github.com/justincbagley/PIrANHA>.
+
+ ${bold}REFERENCES${reset}
+ Bagley, J.C. 2019. PIrANHA v0.4a4. GitHub repository, Available at:
+	<https://github.com/justincbagley/PIrANHA>.
+ Stamatakis, A. 2014. RAxML version 8: a tool for phylogenetic analysis and post-analysis of 
+	large phylogenies. Bioinformatics, 30, 1312-1313.
+
+ Created by Justin Bagley on Wed, Mar 6 09:57:26 CST 2019.
+ Copyright (c) 2019 Justin C. Bagley. All rights reserved.
+"
+
+if [[ "$1" == "-h" ]] || [[ "$1" == "--help" ]]; then
+	echo "$USAGE"
+	exit
+fi
+
+if [[ "$1" == "-V" ]] || [[ "$1" == "--version" ]]; then
+	echo "$(basename "$0") $VERSION";
+	exit
+fi
+
+
+# ############# ############# #############
+# ##       TIME TO RUN THE SCRIPT        ##
+# ##                                     ##
+# ## You shouldn't need to edit anything ##
+# ## beneath this line                   ##
+# ##                                     ##
+# ############# ############# #############
+
+# Trap bad exits with your cleanup function
+trap trapCleanup EXIT INT TERM
+
+# Set IFS to preferred implementation
+IFS=$'\n\t'
+
+# Exit on error. Append '||true' when you run the script if you expect an error.
+set -o errexit
+
+# Run in debug mode, if set
+if ${debug}; then set -x ; fi
+
+# Exit on empty variable
+if ${strict}; then set -o nounset ; fi
+
+# Bash will remember & return the highest exitcode in a chain of pipes.
+# This way you can catch the error in case mysqldump fails in `mysqldump |gzip`, for example.
+set -o pipefail
+
+# Invoke the checkDependenices function to test for Bash packages.  Uncomment if needed.
+# checkDependencies
+
+# Run the script
+RAxMLRunChecker
+
+# Exit cleanly
+safeExit