miga-base 0.2.0.6

data/scripts/_distances_ref_nomulti.bash

@@ -0,0 +1,105 @@
+#!/bin/bash
+# Available variables: $PROJECT, $DATASET, $RUNTYPE, $MIGA, $CORES, $TMPDIR,
+# 			$NOMULTI, $REF
+
+function checkpoint_n {
+   if [[ $N -eq 10 ]] ; then
+      for t in 01.haai 02.aai 03.ani ; do
+         if [[ -s $TMPDIR/$t.db ]] ; then
+	    tab="aai"
+	    [[ "$t" == "03.ani" ]] && tab="ani"
+	    echo "select count(*) from $tab;" \
+	       | sqlite3 $TMPDIR/$t.db\
+	       || exit 1
+	    cp $TMPDIR/$t.db $t/$DATASET.db
+	 fi
+      done
+      N=0
+   fi
+   let N=$N+1
+}
+
+ESS="../07.annotation/01.function/01.essential"
+
+# Initialize temporals
+for t in 01.haai 02.aai 03.ani ; do
+   [[ -s $t/$DATASET.db ]] && cp $t/$DATASET.db $TMPDIR/$t.db
+done
+echo "create table if not exists aai(seq1 varchar(256), seq2 varchar(256)," \
+   "aai float, sd float, n int, omega int);" | sqlite3 $TMPDIR/02.aai.db
+N=1
+
+# Traverse "nearly-half" of the ref-datasets using first-come-first-served
+for i in $(miga list_datasets -P "$PROJECT" --ref --no-multi) ; do
+   echo "=[ $i ]"
+   date "+%Y-%m-%d %H:%M:%S %z"
+   HAAI=""; AAI=""; ANI="";
+   # Check if the i-th dataset is ready
+   [[ -s $ESS/$i.done && -s $ESS/$i.json ]] || continue
+   # Check if this is done (e.g., in a previous failed iteration)
+   AAI=$(echo "select aai from aai where seq1='$DATASET' and seq2='$i';" \
+      | sqlite3 $TMPDIR/02.aai.db || echo "")
+   # Try the other direction
+   if [[ "$AAI" == "" && -s 02.aai/$i.db ]] ; then
+      cp "02.aai/$i.db" "$TMPDIR/$i.db"
+      AAI=$(echo "select aai from aai where seq2='$DATASET' and seq1='$i';" \
+	 | sqlite3 "$TMPDIR/$i.db" || echo "")
+      rm "$TMPDIR/$i.db"
+   fi
+   # Try with hAAI
+   if [[ "$AAI" == "" ]] ; then
+      [[ -e "$TMPDIR/$DATASET.ess.faa" ]] \
+	 || cp $ESS/$DATASET.ess.faa $TMPDIR/$DATASET.ess.faa
+      HAAI=$(aai.rb -1 $TMPDIR/$DATASET.ess.faa -2 $ESS/$i.ess.faa \
+	 -t $CORES -a -n 10 -S $TMPDIR/01.haai.db --name1 $DATASET \
+	 --name2 $i --lookup-first --no-save-rbm || echo "")
+      if [[ "$HAAI" != "" \
+	    && $(perl -MPOSIX -e "print floor $HAAI") -lt 90 ]] ; then
+	 AAI=$(perl -e \
+	    "printf '%f', 100-exp(2.435076 + 0.4275193*log(100-$HAAI))")
+	 echo "insert into aai values('$DATASET','$i','$AAI',0,0,0);" \
+	    | sqlite3 $TMPDIR/02.aai.db
+      fi
+   fi
+   # Try with complete AAI
+   if [[ "$AAI" == "" ]] ; then
+      [[ -e "$TMPDIR/$DATASET.faa" ]] \
+	 || cp ../06.cds/$DATASET.faa $TMPDIR/$DATASET.faa
+      AAI=$(aai.rb -1 $TMPDIR/$DATASET.faa -2 ../06.cds/$i.faa -t $CORES -a \
+	 -S $TMPDIR/02.aai.db --name1 $DATASET --name2 $i --lookup-first \
+	 || echo "")
+   fi
+   date "+%Y-%m-%d %H:%M:%S %z"
+   # Check if ANI is meaningful
+   if [[ -e "../05.assembly/$DATASET.LargeContigs.fna" \
+	 && -e "../05.assembly/$i.LargeContigs.fna" \
+	 && $(perl -MPOSIX -e "print ceil $AAI") -gt 90 ]] ; then
+      # Check if this is done (e.g., in a previous failed iteration)
+      ANI=$(echo "select ani from ani where seq1='$DATASET' and seq2='$i';" \
+	 | sqlite3 $TMPDIR/03.ani.db || echo "")
+      # Try the other direction
+      if [[ "$ANI" == "" && -s 03.ani/$i.db ]] ; then
+	 cp "03.ani/$i.db" "$TMPDIR/$i.db"
+	 ANI=$(echo "select ani from ani" \
+	    "where seq2='$DATASET' and seq1='$i';" \
+	    | sqlite3 "$TMPDIR/$i.db" || echo "")
+	 rm "$TMPDIR/$i.db"
+      fi
+      # Calculate it
+      if [[ "$ANI" == "" ]] ; then
+	 [[ -e "$TMPDIR/$DATASET.LargeContigs.fna" ]] \
+	    || cp ../05.assembly/$DATASET.LargeContigs.fna \
+	       $TMPDIR/$DATASET.LargeContigs.fna
+	 ANI=$(ani.rb -1 $TMPDIR/$DATASET.LargeContigs.fna \
+	    -2 ../05.assembly/$i.LargeContigs.fna -t $CORES \
+	    -S $TMPDIR/03.ani.db -a --name1 $DATASET --name2 $i \
+	    --no-save-regions --no-save-rbm --lookup-first \
+	    || echo "")
+      fi
+   fi
+   echo "$AAI;$ANI"
+   checkpoint_n
+done
+N=10
+checkpoint_n
+

data/scripts/aai_distances.bash

@@ -0,0 +1,40 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/09.distances/02.aai"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.start"
+
+echo -n "" > miga-project.log
+DS=$(miga list_datasets -P "$PROJECT" --ref --no-multi)
+
+# Extract values
+echo "metric a b value sd n omega" | tr " " "\\t" >miga-project.txt
+for i in $DS ; do
+   echo "SELECT CASE WHEN omega!=0 THEN 'AAI' ELSE 'hAAI_AAI' END," \
+      " seq1, seq2, aai, sd, n, omega from aai;" \
+      | sqlite3 "$i.db" | tr "\\|" "\\t" >>miga-project.txt
+   echo "$i" >> miga-project.log
+done
+
+# R-ify
+echo "
+aai <- read.table('miga-project.txt', sep='\\t', h=T);
+save(aai, file='miga-project.Rdata');
+h <- hist(aai[,'value'], breaks=100, plot=FALSE);
+write.table(
+   cbind(h[['breaks']][-length(h[['breaks']])],h[['breaks']][-1],h[['counts']]),
+   file='miga-project.hist', quote=FALSE, sep='\\t',
+   col.names=FALSE, row.names=FALSE);
+" | R --vanilla
+
+# Gzip
+gzip -9 -f miga-project.txt
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.done"
+miga add_result -P "$PROJECT" -r aai_distances
+

data/scripts/ani_distances.bash

@@ -0,0 +1,39 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/09.distances/03.ani"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.start"
+
+echo -n "" > miga-project.log
+DS=$(miga list_datasets -P "$PROJECT" --ref --no-multi)
+
+# Extract values
+echo "metric a b value sd n omega" | tr " " "\\t" >miga-project.txt
+for i in $DS ; do
+   echo "SELECT 'ANI', seq1, seq2, ani, sd, n, omega from ani ;" \
+      | sqlite3 "$i.db" | tr "\\|" "\\t" >>miga-project.txt
+   echo "$i" >> miga-project.log
+done
+
+# R-ify
+echo "
+ani <- read.table('miga-project.txt', sep='\\t', h=T);
+save(ani, file='miga-project.Rdata');
+h <- hist(ani[,'value'], breaks=100, plot=FALSE);
+write.table(
+   cbind(h[['breaks']][-length(h[['breaks']])],h[['breaks']][-1],h[['counts']]),
+   file='miga-project.hist', quote=FALSE, sep='\\t',
+   col.names=FALSE, row.names=FALSE);
+" | R --vanilla
+
+# Gzip
+gzip -9 -f miga-project.txt
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.done"
+miga add_result -P "$PROJECT" -r ani_distances
+

data/scripts/assembly.bash

@@ -0,0 +1,38 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/05.assembly"
+
+b=$DATASET
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.start"
+
+# Assemble
+FA="../04.trimmed_fasta/$DATASET.CoupledReads.fa"
+[[ -e $FA ]] || FA="$FA.gz"
+[[ -e $FA ]] || FA="../04.trimmed_fasta/$DATASET.SingleReads.fa"
+[[ -e $FA ]] || FA="$FA.gz"
+idba_ud --pre_correction -r "$FA" -o "$DATASET" --num_threads "$CORES"
+
+# Clean
+cd $DATASET
+rm kmer graph-*.fa align-* local-contig-*.fa contig-*.fa
+cd ..
+
+# Extract
+if [[ -s $DATASET/scaffold.fa ]] ; then
+   ln -s $DATASET/scaffold.fa $DATASET.AllContigs.fna
+else
+   ln -s $DATASET/contig.fa $DATASET.AllContigs.fna
+fi
+FastA.length.pl $DATASET.AllContigs.fna | awk '$2>=1000{print $1}' \
+   | FastA.filter.pl /dev/stdin $DATASET.AllContigs.fna \
+   > $DATASET.LargeContigs.fna
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.done"
+miga add_result -P "$PROJECT" -D "$DATASET" -r assembly
+

data/scripts/cds.bash

@@ -0,0 +1,45 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/06.cds"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.start"
+GM=$(dirname -- $(which gmhmmp))
+
+# Register key
+if [[ ! -e .gm_key ]] ; then
+   if [[ -e "$GM/gm_key" ]] ; then
+      cp "$GM/gm_key" ".gm_key"
+   elif [[ -e "$GM/gm_key_64" ]] ; then
+      cp "$GM/gm_key_64" ".gm_key"
+   elif [[ -e "$GM/gm_key_32" ]] ; then
+      cp "$GM/gm_key_32" ".gm_key"
+   elif [[ -e "$GM/.gm_key" ]] ; then
+      cp "$GM/.gm_key" ".gm_key"
+   elif [[ -e "$HOME/.gm_key" ]] ; then
+      cp "$HOME/.gm_key" .
+   else
+      echo "Impossible to find MetaGeneMark key, please register your copy" \
+	 "and place the key in '$GM/gm_key'." >&2
+      exit 1
+   fi
+fi
+
+# Run MetaGeneMark
+gmhmmp -a -d -m "$GM/MetaGeneMark_v1.mod" -f G -o "$DATASET.gff2" \
+   "../05.assembly/$DATASET.LargeContigs.fna"
+
+# Extract
+perl "$GM/aa_from_gff.pl" < "$DATASET.gff2" > "$DATASET.faa"
+perl "$GM/nt_from_gff.pl" < "$DATASET.gff2" > "$DATASET.fna"
+
+# Gzip
+gzip -9 -f "$DATASET.gff2"
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.done"
+miga add_result -P "$PROJECT" -D "$DATASET" -r cds
+

data/scripts/clade_finding.bash

@@ -0,0 +1,27 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/10.clades/01.find"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.start"
+
+# Markov-cluster genomes by ANI
+gunzip -c ../../09.distances/03.ani/miga-project.txt.gz | tail -n+2 \
+   | awk -F"\\t" '{print $2"'"\\t"'"$3"'"\\t"'"$4}' > genome-genome.aai90.rbm
+ogs.mcl.rb -d . -o miga-project.ani-clades -t "$CORES" -i \
+   -f "(\\S+)-(\\S+)\\.aai90\\.rbm"
+cat genome-genome.aai90.rbm | awk -F"\\t" '$3>=95' > genome-genome.ani95.rbm
+ogs.mcl.rb -d . -o miga-project.ani95-clades -t "$CORES" -b \
+   -f "(\\S+)-(\\S+)\\.ani95\\.rbm"
+
+# Propose clade projects
+cat miga-project.ani95-clades | tail -n +2 | tr "," "\\t" | awk 'NF >= 5' \
+   > miga-project.proposed-clades
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.done"
+miga add_result -P "$PROJECT" -r clade_finding
+

data/scripts/distances.bash

@@ -0,0 +1,30 @@
+#!/bin/bash
+# Available variables: $PROJECT, $DATASET, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/09.distances"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.start"
+TMPDIR=$(mktemp -d /tmp/MiGA.XXXXXXXXXXXX)
+trap "rm -rf $TMPDIR; exit" SIGHUP SIGINT SIGTERM
+
+# Check type of dataset
+NOMULTI=$(miga list_datasets -P "$PROJECT" -D "$DATASET" --no-multi \
+   | wc -l | awk '{print $1}')
+REF=$(miga list_datasets -P "$PROJECT" -D "$DATASET" --ref \
+   | wc -l | awk '{print $1}')
+
+# Call submodule
+if [[ "$NOMULTI" -eq "1" && "$REF" -eq "1" ]] ; then
+   source "$MIGA/scripts/_distances_ref_nomulti.bash"
+elif [[ "$NOMULTI" -eq "1" ]] ; then
+   source "$MIGA/scripts/_distances_noref_nomulti.bash"
+fi
+
+# Finalize
+rm -R $TMPDIR
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.done"
+miga add_result -P "$PROJECT" -D "$DATASET" -r distances
+

data/scripts/essential_genes.bash

@@ -0,0 +1,29 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/07.annotation/01.function/01.essential"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.start"
+
+# Find and extract essential genes
+[[ -d "$DATASET.ess" ]] && rm -R "$DATASET.ess"
+mkdir "$DATASET.ess"
+TYPE=$(miga list_datasets -P "$PROJECT" -D "$DATASET" \
+   --metadata "type" | awk '{print $2}')
+if [[ "$TYPE" == "metagenome" || "$TYPE" == "virome" ]] ; then
+   HMM.essential.rb -i "../../../06.cds/$DATASET.faa" -o "$DATASET.ess.faa" \
+      -m "$DATASET.ess/" -t "$CORES" -r "$DATASET" --metagenome \
+      > "$DATASET.ess/log"
+else
+   HMM.essential.rb -i "../../../06.cds/$DATASET.faa" -o "$DATASET.ess.faa" \
+      -m "$DATASET.ess/" -t "$CORES" -r "$DATASET" \
+      > "$DATASET.ess/log"
+fi
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "$DATASET.done"
+miga add_result -P "$PROJECT" -D "$DATASET" -r essential
+

data/scripts/haai_distances.bash

@@ -0,0 +1,39 @@
+#!/bin/bash
+# Available variables: $PROJECT, $RUNTYPE, $MIGA, $CORES
+echo "MiGA: $MIGA"
+echo "Project: $PROJECT"
+source "$MIGA/scripts/miga.bash" || exit 1
+cd "$PROJECT/data/09.distances/01.haai"
+
+# Initialize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.start"
+
+echo -n "" > miga-project.log
+DS=$(miga list_datasets -P "$PROJECT" --ref --no-multi)
+
+# Extract values
+echo "metric a b value sd n omega" | tr " " "\\t" >miga-project.txt
+for i in $DS ; do
+   echo "SELECT 'hAAI', seq1, seq2, aai, sd, n, omega from aai ;" \
+      | sqlite3 "$i.db" | tr "\\|" "\\t" >>miga-project.txt
+   echo "$i" >> miga-project.log
+done
+
+# R-ify
+echo "
+haai <- read.table('miga-project.txt', sep='\\t', h=T);
+save(haai, file='miga-project.Rdata');
+h <- hist(haai[,'value'], breaks=100, plot=FALSE);
+write.table(
+   cbind(h[['breaks']][-length(h[['breaks']])],h[['breaks']][-1],h[['counts']]),
+   file='miga-project.hist', quote=FALSE, sep='\\t',
+   col.names=FALSE, row.names=FALSE);
+" | R --vanilla
+
+# Gzip
+gzip -9 -f miga-project.txt
+
+# Finalize
+date "+%Y-%m-%d %H:%M:%S %z" > "miga-project.done"
+miga add_result -P "$PROJECT" -r haai_distances
+

data/scripts/init.bash

@@ -0,0 +1,211 @@
+#!/bin/bash
+set -e
+
+#=======[ Functions ]
+function ask_user {
+   local question=$1
+   local default=$2
+   echo $question >&2
+   echo -n "  [$default] > " >&2
+   read user_answer
+   user_answer=${user_answer:-$default}
+   echo -n "$user_answer"
+}
+
+function check_req {
+   local bin=$1
+   local default=$(dirname "$(which "$bin")")
+   user_answer=$(ask_user "Where can I find $2 ($3)? $4" "$default")
+   if [[ -x "$user_answer/$bin" ]] ; then
+      export PATH="$PATH:$user_answer"
+      echo "MIGA_PATH=\"$user_answer:\$MIGA_PATH\" # $2" >> "$HOME/.miga_rc"
+   else
+      echo "Cannot find $2 at '$user_answer/$bin'. Aborting..." >&2
+      exit 1
+   fi
+}
+
+function check_rlib {
+   local rlib=$1
+   gotit=$(echo "if(require($rlib)) cat('GOT','IT')" | R --vanilla -q 2>&1 \
+      | grep "GOT IT" | wc -l | awk '{print $1}')
+   [[ "$gotit" == "1" ]]
+}
+
+function check_gem {
+   local gem=$1
+   gotit=$(echo "require '$gem'" | ruby 2>/dev/null && echo 1)
+   [[ "$gotit" == "1" ]]
+}
+
+#=======[ Main ]
+MIGA_STARTUP="no"
+MIGA=$(cd "$(dirname "$0")/.."; pwd)
+echo "
+===[ Welcome to MiGA, the Microbial Genome Atlas ]===
+
+I'm the initialization script, and I'll sniff around your computer to
+make sure you have all the requirements for MiGA Daemons.
+" >&2
+
+if [[ "$(ask_user "Would you like to see all the requirements before starting? (yes / no)" "no")" == "yes" ]] ; then
+   echo "" >&2
+   cat "$MIGA/utils/requirements.txt" >&2
+   echo "" >&2
+fi
+
+if [[ -e "$HOME/.miga_rc" ]] ; then
+   case "$(ask_user "I found a previous configuration. Do you want to load the defaults within? (yes / no / cancel)" "yes")" in
+      yes)
+	 source "$HOME/.miga_rc"
+	 if [[ "$MIGA_CONFIG_DATE" == "" ]] ; then
+	    echo "  Loaded incomplete configuration" >&2
+	 else
+	    echo "  Loaded configuration from $MIGA_CONFIG_DATE" >&2
+	 fi
+	 ;;
+      no)
+	 rm "$HOME/.miga_rc"
+	 ;;
+      cancel)
+	 exit 0
+	 ;;
+      *)
+	 echo "Cannot understand your answer, please use 'yes', 'no', or 'cancel'. Aborting..." >&2
+	 exit 1
+   esac
+fi
+
+echo "#!/bin/bash
+# MiGA made this on $(date)
+" > "$HOME/.miga_rc"
+
+# Check Software requirements
+MIGA_STARTUP=$(ask_user "Is there a script I need to load at startup? (no / path to the script to load)" "$MIGA_STARTUP")
+if [[ "$MIGA_STARTUP" != "no" ]] ; then
+   echo "MIGA_STARTUP='$MIGA_STARTUP'
+source \"\$MIGA_STARTUP\"
+" >> "$HOME/.miga_rc";
+   source "$MIGA_STARTUP";
+fi
+echo "
+Looking for Software requirements:" >&2
+reqs=$(tail -n+3 "$MIGA/utils/requirements.txt" | perl -pe 's/\t+/\t/g')
+IFS_BU=$IFS
+IFS=$'\n'
+for ln in $reqs ; do
+   rname=$(echo "$ln" | awk -F'\t' '{print $1}')
+   rtest=$(echo "$ln" | awk -F'\t' '{print $2}')
+   rwebs=$(echo "$ln" | awk -F'\t' '{print $3}')
+   rhint=$(echo "$ln" | awk -F'\t' '{print $4}')
+   check_req "$rtest" "$rname" "$rwebs" "$rhint"
+done
+IFS=$IFS_BU
+echo "export PATH=\$MIGA_PATH\$PATH" >> "$HOME/.miga_rc"
+
+# Check for R packages
+echo "
+Looking for R packages:" >&2
+if ! check_rlib enveomics.R ; then
+   echo "+ Installing enveomics.R" >&2
+   R CMD INSTALL $(dirname "$(which "FastQ.tag.rb")")/../enveomics.R
+fi
+RLIBS="ape ggdendro ggplot2 gridExtra cluster dendextend vegan scatterplot3d"
+for lib in $RLIBS ; do
+   if ! check_rlib $lib ; then
+      echo "+ Installing $lib" >&2
+      echo "install.packages('$lib', repos='http://cran.rstudio.com/')" \
+	 | R --vanilla -q
+   fi
+done
+
+# Check for ruby gems
+echo "
+Looking for Ruby gems:" >&2
+GEMS="rest_client sqlite3 daemons json"
+for gem in $GEMS ; do
+   if ! check_gem $gem ; then
+      echo "+ Installing $gem (user-only)" >&2
+      gem install --user $gem
+   fi
+done
+
+# Check for other files
+echo "
+Looking for additional files:
++ MetaGeneMark license key" >&2
+GM=$(dirname -- $(which gmhmmp))
+if [[ ! -e "$GM/gm_key" && ! -e "$GM/gm_key_64" && ! -e "$GM/gm_key_32" && ! -e "$GM/.gm_key" && ! -e "$HOME/.gm_key" ]] ; then
+   echo "Cannot find it, please place your license key in '$GM/gm_key'. Aborting..." >&2
+   exit 1
+fi
+echo "+ MetaGeneMark scripts" >&2
+if [[ ! -e "$GM/aa_from_gff.pl" || ! -e "$GM/nt_from_gff.pl" ]] ; then
+   echo "Cannot find it, please place aa_from_gff.pl and nt_from_gff.pl in '$GM/'. Aborting..." >&2
+   exit 1
+fi
+echo "+ MyTaxa scores database" >&2
+MT=$(dirname -- $(which MyTaxa))
+if [[ ! -d "$MT/db" ]] ; then
+   echo "Cannot find it, please execute 'python $MT/utils/download_db.py'. Aborting..." >&2
+   exit 1
+fi
+echo "+ MyTaxa DIAMOND database" >&2
+if [[ ! -e "$MT/AllGenomes.faa.dmnd" ]] ; then
+   echo "Cannot find it, please download 'http://enve-omics.ce.gatech.edu/data/public_mytaxa/AllGenomes.faa.dmnd' into '$MT'. Aborting..." >&2
+   exit 1
+fi
+
+# Configure daemon
+echo "
+Default daemon configuration:" >&2
+dtype=$(ask_user "Please select the type of daemon you want to setup (bash / qsub / msub)", "bash")
+case "$dtype" in
+   bash)
+      dlatency=$(ask_user "For how long should I sleep? (# in seconds)" "30")
+      dmaxjobs=$(ask_user "How many jobs can I launch at once?" "6")
+      dppn=$(ask_user "How many CPUs can I use per job?" "2")
+      echo "Setting up internal daemon defaults, if you don't understand this just leave default values:" >&2
+      dcmd=$(ask_user "How should I launch tasks? Use %1\$s for script path, %2\$s for variables, %3\$d for CPUs, %4\$s for log file, and %5\$s for task name." "%2\$s '%1\$s' &> '%4\$s'")
+      dvar=$(ask_user "How should I pass variables? Use %1\$s for keys and %2\$s for values." "%1\$s=%2\$s")
+      dsep=$(ask_user "What should I use to separate variables?" " ")
+      dalive=$(ask_user "How can I know that a process is still alive? Use %1\$s for PID, output should be 1 for running and 0 for non-running." "ps -p '%1\$s'|tail -n+2|wc -l|awk '{print \$1}'")
+      ;;
+   [qm]sub)
+      dqueue=$(ask_user "What's the name of the queue I should use?" "")
+      dlatency=$(ask_user "How long should I sleep? (# in seconds)" "150")
+      dmaxjobs=$(ask_user "How many jobs can I launch at once?" "300")
+      dppn=$(ask_user "How many CPUs can I use per job?" "4")
+      echo "Setting up internal daemon defaults, if you don't understand this just leave default values:" >&2
+      dcmd=$(ask_user "How should I launch tasks? Use %1\$s for script path, %2\$s for variables, and %3\$d for CPUs, %4\$d for log file, and %5\$s for task name." \
+	 "$dtype -q '$dqueue' -v '%2\$s' -l nodes=1:ppn=%3\$d %1\$s -j oe -o '%4\$s' -N '%5\$s' | grep .")
+      dvar=$(ask_user "How should I pass variables? Use %1\$s for keys and %2\$s for values." "%1\$s=%2\$s")
+      dsep=$(ask_user "What should I use to separate variables?" ",")
+      if [[ "$dtype" == "qsub" ]] ; then
+	 dalive=$(ask_user "How can I know that a process is still alive? Use %1\$s for job id, output should be 1 for running and 0 for non-running." \
+	    "qstat -f '%1\$s'|grep ' job_state ='|perl -pe 's/.*= //'|grep '[^C]'|tail -n1|wc -l|awk '{print \$1}'")
+      else
+	 dalive=$(ask_user "How can I know that a process is still alive? Use %1\$s for job id, output should be 1 for running and 0 for non-running." \
+	    "checkjob '%1\$s'|grep '^State:'|perl -pe 's/.*: //'|grep 'Deferred\\|Hold\\|Idle\\|Starting\\|Running\\|Blocked'|tail -n1|wc -l|awk '{print \$1}'")
+      fi
+      ;;
+   *)
+esac
+echo "{
+   \"created\": \"$(date "+%Y-%m-%d %H:%M:%S %z")\",
+   \"updated\": \"$(date "+%Y-%m-%d %H:%M:%S %z")\",
+   \"type\"   : \"$dtype\",
+   \"cmd\"    : \"$dcmd\",
+   \"var\"    : \"$dvar\",
+   \"varsep\" : \"$dsep\",
+   \"alive\"  : \"$dalive\",
+   \"latency\": $dlatency,
+   \"maxjobs\": $dmaxjobs,
+   \"ppn\"    : $dppn
+}" > $HOME/.miga_daemon.json
+
+# Confirm configuration
+echo "
+MIGA_CONFIG_DATE='$(date "+%Y-%m-%d %H:%M:%S %z")'
+" >> "$HOME/.miga_rc"
+

data/scripts/miga.bash

@@ -0,0 +1,12 @@
+#!/bin/bash
+set -e
+#MIGA=${MIGA:-$(cd "$(dirname "$0")/.."; pwd)}
+source "$HOME/.miga_rc"
+export PATH="$MIGA/bin:$PATH"
+
+function exists { [[ -e "$1" ]] ; }
+
+#if [[ "$RUNTYPE" == "qsub" ]] ; then
+#elif [[ "$RUNTYPE" == "msub" ]] ; then
+#fi
+