Shell Programming and Scripting

#!/bin/bash
# README(v1) 12_16_2015

echo " Welcome to NGS module (v1) for medical exome analysis, the following steps will be performed
        ==================================
        1  remove @PG tag
        2  run picard markduplicates
        3  run picard HSmeterics
        4  run bedTools coverageBed
        5  detect bases less than 30 reads
        6  combine files
        =================================="
        
printf "\n\n"
printf "What is the date to analyze : "; read id
    date=$(date -d "$id" "+%m-%d-%Y") #convert to month-date-year
    [ -z "$id" ] && printf "\n No date supplied. Leaving match function." && sleep 2 && printf "\n Goodbye! " && exit
        for f in /home/cmccabe/Desktop/NGS/bam/$date/*.bam ; do
        bname=${f##*/}
        pref=${bname%%.bam}
        echo "detected ${pref}.bam in the directory"
done
printf "Are all the files there? "; read match_choice
case "$match_choice" in
        [yY]) remove ;;
        [nN]) printf "Not all the files are there, please find the missing files(s) and try again " && sleep 2 && printf "\n Goodbye! " && exit
        esac
remove() {
for f in /home/cmccabe/Desktop/NGS/bam/$date/*.bam ; do
    bname=`basename $f`
    pref=${bname%%.bam}
    samtools view -H $f | sed '/^@PG/d' | samtools reheader - $f > home/cmccabe/Desktop/NGS/bam/$date/${pref}_newheader.bam
done
}

#!/bin/bash
#README(v1.1) 12_16_2015

############ Functions ############

# Enter a comment to why this function was coded.
# Avoid to comment what the function is doing.
# It is more important to know the why that the what.
# Some comment examples are given below, highlighted in blue.
remove() {

    for f in /home/cmccabe/Desktop/NGS/bam/$date/*.bam; do
        bname="${f##*/}"
        pref="${bname%%.bam}"
        samtools view -H "$f" |
                sed '/^@PG/d' |
                samtools reheader - "$f" > "home/cmccabe/Desktop/NGS/bam/${date}/${pref}_newheader.bam"
    done
}

###################################

# It is more proper to end printf with the right amount of new lines
# you want to format, that to add it to the beginning of a string
# in the next printf

printf " Welcome to NGS module (v1) for medical exome analysis, the following steps will be performed
        ==================================
        1  remove @PG tag
        2  run picard markduplicates
        3  run picard HSmeterics
        4  run bedTools coverageBed
        5  detect bases less than 30 reads
        6  combine files
        ==================================\n\n"

# It would be nice to give the user a visual clue
# of how the date format looks like
printf "What is the date to analyze?: "
read id

# bail out if a date is not given
[ -z $id ] && printf "No date was supplied. This program is existing...\n" \
           && sleep 2                                                      \
           && printf "Goodbye!\n"                                          \
           && exit
# now you or others might know that the idea was to bail out if a date was
# not given, but easy a bug since entering any
# character is still accepted as a date.

# dates must be formatted to match directory name form
date=$(date -d "$id" "+%m-%d-%Y")

# user need to visually verify the files outputed and confirm
for f in /home/cmccabe/Desktop/NGS/bam/$date/*.bam; do
    bname="${f##*/}"
    pref="${bname%%.bam}"
    echo "detected ${pref}.bam in the directory"
done

printf "Are all the files there? Please ENTER (Y/N): "
read match_choice

# The user must confirm by entering either Y or y, n or N
# Again, commenting the why gives an idea of the intention for this code,
# making it easy to debug when the user enters a "don't know" or just
# presses ENTER.
case "$match_choice" in
        [yY]) remove ;;
        [nN]) printf "Not all the files were there."
              printf "Please find the missing files(s) and try again\n"
              sleep 2
              printf "Goodbye!"
              exit ;;
esac