extractSampleDates.sh

#!/bin/bash

### This is preparatory for evaluateTimings.R; run this on an RV217 or CAP/V3 dir to extract the dates from the file names, like so:
# ./evaluateTimings.sh /fh/fast/edlefsen_p/bakeoff/analysis_sequences/raw/nflg/1m/ > /fh/fast/edlefsen_p/bakeoff_analysis_results/raw/nflg/1m/sampleDates.tbl
# ./evaluateTimings.sh /fh/fast/edlefsen_p/bakeoff/analysis_sequences/raw/v3/1m/ > /fh/fast/edlefsen_p/bakeoff_analysis_results/raw/v3/1m/sampleDates.tbl

# Note that if you provide a second argument, it will look for "list files" called eg nnnnnnn.list instead of eg processed_nnnnnnn.list.
## SEE NOTE BELOW REGARDING VARYING DATES: USES DATE OF FINAL FASTA ENTRY IN EACH INPUT FILE!
export mainDir=$1

if [ -z $2 ]; then
    export listPrefix="processed_"
else
    export listPrefix=""
fi

for patient in  `ls -c1 ${1}/*.list  | egrep --only "[0-9]+\.list" | egrep --only "[0-9]+"  | sort -u`
do
    export listFile=${mainDir}/${listPrefix}${patient}.list
    
    export fastaFile=`head -n 1 ${listFile} | sed "s/\/home\/tholzman\/edf/\/fh\/fast\/edlefsen_p/g"`
    #echo ${patient} ${fastaFile}

    ## NOTE that this uses the last entry in the file to extract the date.  This is done to ensure that if there are multiple dates in the file, only the latest will be used -- but THIS ASSUMES THAT THE DATES ARE IN ORDER.
    echo ${patient} `grep "^>" ${fastaFile} | tail -n 1  | head -n 1 | cut -d '|' -f 5`
done