reformat-seq.sh

mfold

#!/bin/sh

# reformat-seq.sh
# Input: Sequence file $1
# FILE_PREFIX: The sequence file name is stripped of .seq, .fasta, etc.
# Output file is $FILE_PREFIX-local.seq
# Cases:

# IG (Intelligenetics) is left as is an the output file is just a
# symbolic link to the input file.

# FASTA is left as is but the header line is trucated at 80 characters
# and the sequence itself has all white space removed and is wrapped
# to 75 nt per record. 

# Both GenBank and EMBL are parsed properly. White spaces are
# compressed to single spaces. The output is is FASTA format.

# GCG is no longer accepted.

# If the file does not appear to be in IG, FASTA, GB or EMBL format,
# then it is treated as raw sequence data. The internal name for the
# sequence is $FILE_PREFIX and the output is in FASTA format.

if [ $# -lt 1 ] ; then
	echo -e " *** Usage: reformat-seq.sh in_fil,\nwhere in_fil is the name of the input sequence file."
	exit 1
elif [ ! -s $1 ] ; then
    echo -e "The file, $1, does not exist or is empty"
    exit 2
fi

FILE_PREFIX=`echo $1|sed 's/.seq//'|sed 's/.gb//'|sed 's/.embl//'|sed \
    's/.fasta//'|sed 's/.SEQ//'|sed 's/.GB//'|sed 's/.EMBL//'|sed 's/.FASTA//'`

#Test for GenBank
GB=`grep -n "^LOCUS " $1`
if [ $? = 0 ] ; then
    NAME_INDEX=`echo $GB|cut -d: -f1`
    NAME=`tail +$NAME_INDEX $1|head -1|tr -s " " " "|sed 's/^ //'|cut -c1-72`
    START=`grep -n "^ORIGIN " $1`
    if [ $? = 0 ] ; then
	START=`echo $START|cut -d: -f1`
	START=`expr $START + 1`
    else
	echo -e "Corrupted GenBank file. No ORIGIN line found."
	exit 3
    fi
    STOP=`grep -n "^//" $1`
    if [ ! $? = 0 ] ; then
	STOP=`wc $i|tr -s " " " "|sed 's/^ //'`
    else
	STOP=`echo $STOP|cut -d: -f1`
    fi

    echo -e ">$NAME" > ${FILE_PREFIX}-local.seq
    STOP=`expr $STOP - 1`
    NLINES=`expr $STOP - $START + 1`
#    tail +$START $1|tr -cd "\012 ,A-z"|tr "a-z" "A-Z"|tr -s " " " "|\
#	head -$NLINES >> ${FILE_PREFIX}-local.seq
    tail +$START $1|tr -cd "\012 ,A-z"|tr -s " " " "|head -$NLINES \
	>> ${FILE_PREFIX}-local.seq
    echo -e "$FILE_PREFIX"
    exit 0
fi

#Test for EMBL
EMBL=`grep -n "^ID " $1`
if [ $? = 0 ] ; then
    NAME_INDEX=`echo $EMBL|cut -d: -f1`
    NAME=`tail +$NAME_INDEX $1|head -1|tr -s " " " "|sed 's/^ //'|cut -c1-72`
    START=`grep -n "^SQ " $1`
    if [ $? = 0 ] ; then
	START=`echo $START|cut -d: -f1`
	START=`expr $START + 1`
    else
	echo -e "Corrupted EMBL file. No SQ line found."
	exit 3
    fi
    STOP=`grep -n "^//" $1`
    if [ ! $? = 0 ] ; then
	STOP=`wc $i|tr -s " " " "|sed 's/^ //'`
    else
	STOP=`echo $STOP|cut -d: -f1`
    fi

    echo -e ">$NAME" > ${FILE_PREFIX}-local.seq
    STOP=`expr $STOP - 1`
    NLINES=`expr $STOP - $START + 1`
#    tail +$START $1|tr -cd "\012 ,A-z"|tr "a-z" "A-Z"|tr -s " " " "|\
#	head -$NLINES >> ${FILE_PREFIX}-local.seq
    tail +$START $1|tr -cd "\012 ,A-z"|tr -s " " " "|head -$NLINES \
	>> ${FILE_PREFIX}-local.seq
    echo -e "$FILE_PREFIX"
    exit 0
fi

#Test for FASTA
FASTA=`head -1 $1|grep "^>"`
if [ $? = 0 ] ; then
    NAME=`echo $FASTA|cut -c2-80`
    echo -e ">$NAME" > ${FILE_PREFIX}-local.seq
#    tail +2 $1|tr -cd "A-z"|tr "a-z" "A-Z"|fold -w 75\
#	>> ${FILE_PREFIX}-local.seq
    tail +2 $1|tr -cd "A-z"|fold -w 75 >> ${FILE_PREFIX}-local.seq
    echo '' >> ${FILE_PREFIX}-local.seq
    echo -e "$FILE_PREFIX"
    exit 0
fi

#Test for IG
IG=`head -1 $1|grep "^;"`
if [ $? = 0 ] ; then
    rm -f ${FILE_PREFIX}-local.seq
    ln -s $1 ${FILE_PREFIX}-local.seq
    echo -e "$FILE_PREFIX"
    exit 0
fi

#At this point, assume raw sequence data and hope for the best
echo -e ">$FILE_PREFIX" > ${FILE_PREFIX}-local.seq
#cat $1|tr -cd "A-z"|tr "a-z" "A-Z"|fold -w 75 >> ${FILE_PREFIX}-local.seq
cat $1|tr -cd "A-z"|fold -w 75 >> ${FILE_PREFIX}-local.seq
echo '' >> ${FILE_PREFIX}-local.seq
echo -e "$FILE_PREFIX"
exit 0