Commit 863c6d2e authored by kbattenb's avatar kbattenb
Browse files

convert.sh syntax now internally consistent

parent 465e29dd

launch_universc.sh

+213 −248
Original line number Diff line number Diff line
@@ -8,12 +8,14 @@ if [[ -z $cellrangerpass ]]; then 
    echo "cellranger command is not found."

    exit 1

fi

ver_info=`paste -d "\n" <(cellranger count --version) <(echo conversion script version 0.2.0.9001) | head -n 3 | tail -n 2`

ver_info=`paste -d "\n" <(cellranger count --version) <(echo conversion script version 0.2.0.9002) | head -n 3 | tail -n 2`

##########



#####locate script for importing barcodes######





#####locate launch_universc.sh for importing barcodes######

SOURCE="${BASH_SOURCE[0]}"

while [ -h "$SOURCE" ]; do # resolve $SOURCE until the file is no longer a symlink

while [[ -h "$SOURCE" ]]; do #resolve $SOURCE until the file is no longer a symlink

    TARGET="$(readlink "$SOURCE")"

    if [[ $TARGET == /* ]]; then

        echo "SOURCE '$SOURCE' is an absolute symlink to '$TARGET'"
@@ -21,16 +23,18 @@ while [ -h "$SOURCE" ]; do # resolve $SOURCE until the file is no longer a symli 
    else

        SCRIPT_DIR="$( dirname "$SOURCE" )"

        echo "SOURCE '$SOURCE' is a relative symlink to '$TARGET' (relative to '$SCRIPT_DIR')"

    SOURCE="$SCRIPT_DIR/$TARGET" # if $SOURCE was a relative symlink, we need to resolve it relative to the path where the symlink file was located

        SOURCE="$SCRIPT_DIR/$TARGET" #if $SOURCE is a relative symlink, we need to resolve it relative to the path where the symlink file was located

    fi

done

echo "SOURCE is '$SOURCE'"

RDIR="$( dirname "$SOURCE" )"

SCRIPT_DIR="$( cd -P "$( dirname "$SOURCE" )" >/dev/null 2>&1 && pwd )"

if [ "$DIR" != "$RDIR" ]; then

if [[ $RDIR != $SCRIPT_DIR ]]; then

    echo "DIR '$RDIR' resolves to '$SCRIPT_DIR'"

fi

echo "Running convertion script in '$SCRIPT_DIR'"

echo "Running launch_universc.sh in '$SCRIPT_DIR'"

##########







#####usage statement#####

help='
@@ -62,7 +66,7 @@ Mandatory arguments to long options are mandatory for short options too. 
  -v,  --version                Output version information and exit

       --verbose                Print additional outputs for debugging



For each fastq file, follow the following naming convention:

For each fastq file, follow the naming convention below:

  <SampleName>_<SampleNumber>_<LaneNumber>_<ReadNumber>_001.fastq

  e.g. EXAMPLE_S1_L001_R1_001.fastq

       Example_S4_L002_R2_001.fastq.gz
@@ -266,60 +270,54 @@ done 


#####check if UniverSC is running already#####

#create .lock file if none exists

if [[ ! -f  ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock ]]

    then

if [[ ! -f  ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock ]]; then

    echo "creating lock file"

    echo 0 > ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock

fi

#import lock counter



#check if jobs running (check value in .lock file)

echo "checking .lock file"

lock=`cat ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock`

#check if jobs running

if [[ ! $lock == "0" ]]

    then

    echo "$lock number of cellranger ${VERSION} jobs running in ${DIR}"



if [[ ! $lock == "0" ]]; then

    echo " $lock number of cellranger ${VERSION} jobs already running in ${DIR}"

    #check technology current running

    if [[ -f ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called ]]

        then

    if [[ -f ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called ]]; then

        last=`cat ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called`

        echo "running technology $last with $lock jobs"

        #check if currently running technology is different to convert call

        if [[ $last == $technology ]]

           then

        echo " running $lock jobs with technology $last"

        #check if the technology running is different from the current convert call

        if [[ $last == $technology ]]; then

            echo " no conflict detected"

           #add disable increment for setup calls (which aren't counted or removed)

           if [[ $setup == false ]]

               then

            #add disable increment for setup calls (which are not counted or removed)

            if [[ $setup == false ]]; then

                #add current job to lock

                echo " increment lock"

                lock=$(($lock+1))

                echo $lock > ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock

                echo " call accepted: running $lock cellranger jobs on $technology"

            fi

           echo "call accepted: running $lock cellranger calls on $technology"

	else

           echo "conflict between $technology and current $lock cellranger runs on $last"

           echo "***Please hold calls for $technology until jobs running $last are completed"

           echo "***Warning: remove ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock if $last jobs have completed or aborted"

           echo "***Error: barcode whitelist configured for currently running technology: $last" 

	    echo "Error: conflict between technology selected for the new job ($technology) and for $lock jobs currently running ($last)"

            echo "barcode whitelist configured and locked for currently running technology: $last"

            echo "remove ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock if $last jobs have completed or aborted"

            exit 1

        fi

    fi

else

    ## initialise lock file if first call (no other jobs running)

    #add disable increment for setup calls (which aren't counted or removed)

    if [[ $setup == false ]]

        then

    #initialise lock file if first call (no other jobs running)

    #add disable increment for setup calls (which are not counted or removed)

    if [[ $setup == false ]]; then

        #add current job to lock

        echo " increment lock"

        lock=$(($lock+1))

        echo "no other jobs running: $lock initiated for $technology"

        echo " call accepted: running no other cellranger jobs"

        echo $lock > ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock

    fi

fi



##########



#####check if input maches expected inputs#####

if [[ $verbose == "true" ]]

    then

if [[ $verbose == "true" ]]; then

    echo "checking options ..."

fi
@@ -346,95 +344,70 @@ fi 


#check for file type (extension) for files

##allows incomplete file names and processing compressed files

for i in ${!read1[@]}

do

for i in ${!read1[@]}; do

    read=${read1[$i]}

    if [[ $verbose == "true" ]];

        then

    if [[ $verbose == "true" ]]; then

        echo " checking file format for $read1 ..."

    fi

    if [ -f $read ] && [ ! -h $read ]

        then

        if [[ $read != *"gz" ]]

            then

    if [[ -f $read ]] && [[ -h $read ]]; then

        if [[ $read == *"gz" ]]; then

            gunzip -k $read

            #update file variable

            read=`echo $read | sed -e "s/\.gz//g"`

        fi

        if [[ $read != *"fastq" ]] || [[ $read != *"fq" ]]

            then

            echo "Warning: file $read expected to be in fastq format"

        if [[ $read != *"fastq" ]] && [[ $read != *"fq" ]]; then

            echo "***Warning: file $read is assubed to be in fastq format***"

        fi

        echo $read

    elif [ -f ${read}.fq ] && [ ! -h $read ]

    then

        read=${read}.fq

        echo $read

    elif [ -f ${read}.fastq ] && [ ! -h $read ]

        then

        read=${read}.fastq

        echo $read

    elif [ -f ${read}.fq.gz ] && [ ! -h $read ]

        then

        gunzip -k ${read}.fq.gz

        read=${read}.fq

        echo $read

    elif [ -f ${read}.fastq.gz ] && [ ! -h $read ]

        then

        gunzip -k ${read}.fastq.gz

        read=${read}.fastq

        echo $read

        echo "  $read"

    elif [[ -f $read ]]; then

        if [[ $read == *"gz" ]]; then

            gunzip -k $read

            #update file variable

            read=`echo $read | sed -e "s/\.gz//g"`

        fi

        if [[ $read != *"fastq" ]] && [[ $read != *"fq" ]]; then

            echo "***Warning: file $read is assubed to be in fastq format***"

        fi

        echo "  $read"

    else

        echo $read not found

        echo "Error: $read not found"

        exit 1

    fi

    read1[$i]=$read

done



for i in ${!read2[@]}

do

for i in ${!read2[@]}; do

    read=${read2[$i]}

    if [[ $verbose == "true" ]];

        then

    if [[ $verbose == "true" ]]; then

        echo " checking file format for $read2 ..."

    fi

    if [ -f $read ] && [ ! -h $read ]

        then

        if [[ $read != *"gz" ]]

            then

    if [[ -f $read ]] && [[ -h $read ]]; then

        if [[ $read == *"gz" ]]; then

            gunzip -k $read

            #update file variable

            read=`echo $read | sed -e "s/\.gz//g"`

        fi

        if [[ $read != *"fastq" ]] || [[ $read != *"fq" ]]

            then

            echo "Warning: file $read expected to be in fastq format"

        if [[ $read != *"fastq" ]] && [[ $read != *"fq" ]]; then

            echo "***Warning: file $read is assubed to be in fastq format***"

        fi

        echo $read

    elif [ -f ${read}.fq ] && [ ! -h $read ]

    then

        read=${read}.fq

        echo $read

    elif [ -f ${read}.fastq ] && [ ! -h $read ]

        then

        read=${read}.fastq

        echo $read

    elif [ -f ${read}.fq.gz ] && [ ! -h $read ]

        then

        gunzip -k ${read}.fq.gz

        read=${read}.fq

        echo $read

    elif [ -f ${read}.fastq.gz ] && [ ! -h $read ]

        then

        gunzip -k ${read}.fastq.gz

        read=${read}.fastq

        echo $read

        echo "  $read"

    elif [[ -f $read ]]; then

        if [[ $read == *"gz" ]]; then

            gunzip -k $read

            #update file variable

            read=`echo $read | sed -e "s/\.gz//g"`

        fi

        if [[ $read != *"fastq" ]] && [[ $read != *"fq" ]]; then

            echo "***Warning: file $read is assubed to be in fastq format***"

        fi

        echo "  $read"

    else

        echo $read not found

        echo "Error: $read not found"

        exit 1

    fi

    read2[$i]=$read

done





#renaming read1 and read 2 files if not compatible with the convention.

if [[ $verbose == "true" ]]; then

    echo " checking file name for $read1 ..."
@@ -444,7 +417,7 @@ for i in ${!read1[@]}; do 
    if [[ -h $read ]]; then

        path=`readlink -f $read`

        if [[ $verbose == "true" ]]; then

            echo " ***Warning: file $read not in current directory. Path to the file captured instead***"

            echo " ***Warning: file $read not in current directory. Path to the file captured instead.***"

            echo "  (file) $read"

            echo "  (path) $path"

        fi
@@ -452,23 +425,21 @@ for i in ${!read1[@]}; do 
    fi

    case $read in

        #check if contains lane before read

        *_L0[0123456789][0123456789]_R[12]*)

        *_L0[0123456789][0123456789]_R1*)

            if [[ $verbose == "true" ]]; then

                echo "  $read compatible with lane"

            fi

        ;;

        *) echo "  converting $read ..."

        *) 

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "   assuming 1 lane if not given"

                echo "***Warning: file $read does not have lane value in its name. Lane 1 is assumed.***"

	        echo "  renaming $read ..."

            fi

            rename "s/_R1/_L001_R1/" $read

            #update file variable

            read=`echo $read | sed -e "s/_R1/_L001_R1/g"`

            read1[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read ..."

        fi

        ;;

    esac

    case $read in
@@ -479,18 +450,16 @@ for i in ${!read1[@]}; do 
            fi

        ;;

        *)

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "  converting $read ..."

                echo "***Warning: file $read does not have sample value in its name. Sample $j is assumed.***"

	        echo "  renaming $read ..."

            fi

        #rename file

	    j=$((${i}+1))

            rename "s/_L0/_S${j}_L0/" $read

            #update file variable

            read=`echo $read | sed -e  "s/_L0/_S${j}_L0/g"`

            read1[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read ..."

        fi

        ;;

    esac

    case $read in
@@ -501,20 +470,19 @@ for i in ${!read1[@]}; do 
            fi

        ;;

        *)

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "  converting $read ..."

                echo "***Warning: file $read does not have suffix in its name. Suffix 001 is given.***"

                echo "  renaming $read ..."

            fi

        #rename file

	    rename "s/_R1.*\./_R1_001\./" $read

            #update file variable

            read=`echo $read | sed -e  "s/_R1.*\./_R1_001\./g"`

            read1[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read"

        fi

        ;;

    esac

done



if [[ $verbose == "true" ]]; then

    echo " checking file name for $read2 ..."

fi
@@ -523,7 +491,7 @@ for i in ${!read2[@]}; do 
    if [[ -h $read ]]; then

        path=`readlink -f $read`

        if [[ $verbose == "true" ]]; then

            echo " ***Warning: file $read not in current directory. Path to the file captured instead***"

            echo " ***Warning: file $read not in current directory. Path to the file captured instead.***"

            echo " (file) $read"

            echo " (path) $path"

        fi
@@ -531,23 +499,21 @@ for i in ${!read2[@]}; do 
    fi

    case $read in

        #check if contains lane before read

        *_L0[0123456789][0123456789]_R[12]*)

        *_L0[0123456789][0123456789]_R2*)

            if [[ $verbose == "true" ]]; then

                echo "  $read compatible with lane"

            fi

        ;;

        *) echo "  converting $read ..."

        *) 

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "   assuming 1 lane if not given"

                echo "***Warning: file $read does not have lane value in its name. Lane 1 is assumed.***"

	        echo "  renaming $read ..."

            fi

            rename "s/_R2/_L001_R2/" $read

            #update file variable

            read=`echo $read | sed -e "s/_R2/_L001_R2/g"`

            read2[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read ..."

        fi

        ;;

    esac

    case $read in
@@ -558,18 +524,16 @@ for i in ${!read2[@]}; do 
            fi

        ;;

        *)

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "  converting $read ..."

                echo "***Warning: file $read does not have sample value in its name. Sample $j is assumed.***"

	        echo "  renaming $read ..."

            fi

        #rename file

	    j=$((${i}+1))

            rename "s/_L0/_S${j}_L0/" $read

            #update file variable

            read=`echo $read | sed -e  "s/_L0/_S${j}_L0/g"`

            read2[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read ..."

        fi

        ;;

    esac

    case $read in
@@ -580,17 +544,15 @@ for i in ${!read2[@]}; do 
            fi

        ;;

        *)

            #rename file

            if [[ $verbose == "true" ]]; then

            echo "  converting $read ..."

                echo "***Warning: file $read does not have suffix in its name. Suffix 001 is given.***"

                echo "  renaming $read ..."

            fi

        #rename file

	    rename "s/_R2.*\./_R2_001\./" $read

            #update file variable

            read=`echo $read | sed -e  "s/_R2.*\./_R2_001\./g"`

            read2[$i]=$read

        if [[ $verbose == "true" ]]; then

            echo "   renaming $read"

        fi

        ;;

    esac

done
@@ -653,7 +615,7 @@ elif [[ $id == *" "* ]]; then 
fi



#check if reference is present

if [[ -z $reference ]] && [[ $setup == "false" ]]; then

if [[ -z $reference ]] && [[ $setup != "false" ]]; then

    echo "Error: option --reference is required"

    exit 1

fi
@@ -693,8 +655,8 @@ fi 




####report inputs#####

echo "

#####Input information#####"

echo ""

echo "#####Input information#####"

echo "SETUP: $setup"

if ! [[ $setup == "false" ]]; then

    echo "***Warning: whitelist is converted for compatibility, valid barcodes cannot be detected accurately with this technology***"
@@ -1099,30 +1061,33 @@ end=`date +%s` 
runtime=$((end-start))

##########







#####remove files if convert is not running elsewhere#####

echo "updating .lock file"



#reset lock counter (read in case changed by other jobs)

lock=`cat ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock`



#remove currewnt job from counter (successfully completed)

lock=$(($lock-1))

#check if jobs running

if [[ $lock -ge 1 ]]

    then

if [[ $lock -ge 1 ]]; then

    echo "$lock number of cellranger ${VERSION} jobs still running in ${DIR}"

    #check technology current running

    if [[ -f ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called ]]

        then

    if [[ -f ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called ]]; then

        last=`cat ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.last_called`

        echo "running technology $last with $lock jobs"

    fi

fi



#remove .lock file if no other jobs running exists (prevents negative values allowing technologies to run at same time)

if [[ $lock -le 0 ]]

    then

if [[ $lock -le 0 ]]; then

    echo "no other jobs currently running: lock files cleared for cellranger ${VERSION} in ${DIR}"

    echo "no conflicts: whitelist can now be changed for other technologies"

    rm -f ${DIR}-cs/${VERSION}/lib/python/cellranger/barcodes/.lock

fi

##########

CRA Git | Maintained and supported by SUSTech CRA and CCSE