#! /usr/bin/awk -f  
# fix_busco_CDS_frame.awk    Bernardo    for Steganinae phylogeny project  9mar2018  v.10mar2018 
# BUSCO produces some CDS  with frame +2 or +3 which created problems in the posterior analysis, which are  ot well solved  by running translatorX with the -t T option. 
 
# uses Busco list of single orthologs , both the CDS (fna) and protein (faa) sequences 
#usage  fix_busco_CDSframe.awk  mode=detect   diptera_odb9.gene_name


function usage(dummy){
    print "BUSCO produce some CDS  with frame +2 or +3 which created problems in the posterior analysis. "
	print "This script will convert all them to frame +1 by adding N or NN at the begining of the sequence, and will overwrite the wrong CDS"
	print "usage:"
	print "fix_busco_CDS_frame.awk   diptera_odb9.gene_name" 
	print "fix_busco_CDS_frame.awk   mode=detect  diptera_odb9.gene_name     [mode=detect will only detect tye probem. mode=fix is the default] "
	print "other options:  verbose=0 [no log is printed]    verbose=1   [prints total CDS found, total CDS corrected; default]     verbose=2   [same as 1 , plus lists each corrected CDS ]"
	flag_exit=1; exit
	}
	
	 
BEGIN{	if (ARGC==1){usage(dummy)} 
		total_CDS=0;  wrong_frame_CDS=0 ; fixed_CDS=0
		}

(NR==1){
		# n_species=split(species,species_array)
		# for (i=1; i<=n_species; i++){output_prefix=output_prefix species_array[i] "_"}   # Ehians_Git3_nav_
		# if (file_extension == ""){file_extension="fna"}
		if (mode == ""){mode="fix"}
		if (verbose == ""){verbose="1"}
		}

(NR>=1){
		current_ortholog=$1
		ortholog_count ++
		flag_files_existence =  system("test -f " current_ortholog".fna")  +  system("test -f " current_ortholog".faa")   #   0 if the expression is true,   1 if the expression is false #it is normal that some files do not exist: the corresponding ortholog was not annotated by BUSCO in the species.
		if (flag_files_existence == 0){
				total_CDS_found ++
				flag_blast_error=0 ; frame=0
				# bl2seq -p blastx -i phoEOG0915074U_CDS.fasta -j phoEOG0915074U_prot.fasta -F F -D 1 -e 0.00001 
				cmd_bl2seq = "bl2seq -p blastx -i " current_ortholog ".fna  -j " current_ortholog ".faa -F F -D 1 -e 0.00001 | sed  '/^#/d' "
				flag_blast_error = cmd_bl2seq | getline bl2seq_output   #return value from getline: 1 if successful, -1 if there is an error
				if (flag_blast_error != 1){print "error while performing command " cmd_bl2seq ;  exit}
				close(cmd_bl2seq) 
				# bl2seq Fields: Query id, Subject id, % identity, alignment length, mismatches, gap openings, q. start, q. end, s. start, s. end, e-value, bit score
				split(bl2seq_output,a)
				identity=a[3]  ; query_start=a[7]
				frame =  (query_start -1 )%3 +1    #alignment start with 1,4,7... frame +1      alignment start with 2,5,10... frame +2      alignment start with 3,6,11... frame +3
				if (frame != 1){
						wrong_frame_CDS ++
						flag_fix_frame_error=0 ;
						if (verbose >=2) {print current_ortholog "CDS is on frame +" frame }
						addedN="nnn"      #just a mark; should not be used
						if (frame==2){addedN="nn"}
						if (frame==3){addedN="n"}				
						cmd_fix_frame =  "sed -e '/>/{s/$/\\n" addedN "/}' "  current_ortholog ".fna "  "> temp.fna ;"   " mv temp.fna " current_ortholog ".fna "
						# print cmd_fix_frame
						if (mode=="fix"){ 
								flag_fix_frame_error = system(cmd_fix_frame) 
								fixed_CDS ++
								if (flag_fix_frame_error == -1){print "error while performing command " cmd_fix_frame ;  exit}
								}  				
						}
				}
		}

		

END{
	if (verbose >=1){print "CDS listed in input file: " ortholog_count, "total CDS found: " total_CDS_found, "wrong frame CDS: "  wrong_frame_CDS, "fixed CDS: ", fixed_CDS	}
	}


	
	# EOG09150006
	# EOG09150009
	# EOG0915000C
	# EOG0915000S
	# EOG0915000X
	# awk 'BEGIN {print getline < "file" < 0 ? "not exists" : "exists"}'
	# if (system("test -f " FILE)) { ... } 
	
	
	# awk 'BEGIN{print (system("test -f " FILE))}'
	 # awk 'BEGIN{print (system("test -f  coloc.data"))}'
	 
	 # awk 'BEGIN {
       # cmd = "date -j -f %s"
       # cmd | getline mydate
       # close(cmd)
     # }'