UNIX for Dummies Questions & Answers

>Sample23
acagtagaca-atagacgatggagagacatgaggcccaaaattt
>Sample-123
--agtagacagatagacgatggagagacatgaggcccaattt--
>Sample23 Freq 45
acagtagacagatagacgat-gagagacatgaggcccaaaattt
>Reference
---------agatagacgatggagagacatgaggccca------
>Sample__23
acagtagacagatagacgatggagagacatgaggcccaaaa---

>Sample23
a-atagacgatggagagacatgaggccca
>Sample-123
agatagacgatggagagacatgaggccca
>Sample23 Freq 45
agatagacgat-gagagacatgaggccca
>Reference
agatagacgatggagagacatgaggccca
>Sample__23
agatagacgatggagagacatgaggccca

awk 'NR==FNR{if($0 ~ /Reference/){getline; gsub("-","");x=length;}next;}{print substr($0,1,x);}' infile infile

>Reference
-ag---atagac---gatgg--agaga-catga--ggccca---

awk '
reffound == 1 {			# We are processing the reference data line...
	match($1, /[^-]+/)	# Find the string of non-hyphens and set the
				# RSTART and RLENGTH variables.
	reffound = 0		# Clear the Reference found flag.
	nextfile		# If your version of awk contains this function,
				# your script will run faster; if it does not
				# support the nextfile function, comment it out
				# or remove these four lines.
}
FNR == NR {			# We are reading the file for the first time...
	if($1 == ">Reference")	# Look for the Reference header...
		reffound = 1	# Found it; set the Reference found flag.
	next			# Skip remaining lines in this script and resume
				# processing with the next input line at the top
				# of the script.
}
				# We are reading the file the 2nd time now...
FNR % 2				# Print header lines (odd line numbers).
(FNR + 1) % 2 {			# Print selected part of data (even) lines.
	print substr($1, RSTART, RLENGTH)
}' infile infile

awk '
  {
    h=$1                                        # set header variable to $1 in the ">" terminated record
    sub(h FS,x)                                 # delete the header and the first newline
                                                # what remains is the sequence data
  } 
  NR==FNR {                                     # if the input file is read for the first time
    if(h=="Reference") {                        # if the reference is found
      match($0, /[^-]+/)                        # determine the relative start and length of the reference sequence
    }
    next                                        # process the next line 
  } 
  FNR>1 {                                       # When the file is read the second line, ignore the first empty record
    print RS h FS substr($0, RSTART, RLENGTH)   # Print RS, the header, a newline and the 
                                                # sequence, trimmed to the position and width from the reference
  }
' RS=\> ORS='\n' FS='\n' infile infile          # read the input twice and set RS to ">"