#! /usr/bin/awk -f  
# busco2multifasta.awk    Bernardo    for Steganinae phylogeny project  16feb2018     v. 10mar2018
# uses Busco list of single orthologs from two or more species to produece a multifasta ready for alignment
#usage  busco2multifasta.awk    species="EhiansMLmf1  Git3  vir"   diptera_odb9.gene_name
#       busco2multifasta.awk    species="EhiansMLmf1  Git3  vir"  file_extension=faa diptera_odb9.gene_name


function usage(dummy){
	print "usage:"
	print "busco2multifasta.awk    species=\42Ehians  Git3  nav\42   diptera_odb9.gene_name"
	print "busco2multifasta.awk    species=\42Ehians  Git3  nav\42  file_extension=faa  diptera_odb9.gene_name"
    print "busco2multifasta.awk    species=\42Ehians  Git3  nav\42  out_dir="/home/tools/busco/dataset_10mar2018/busco_output/multifasta/"   diptera_odb9.gene_name"	
	print "Ehians  Git3  nav are the names os the folders that contain the fasta of single-copy genes annotated by BUSCO, originally stored in the folder xxxx"
	print "this awk program produce  multifasta files, one for each ortholog, containing the orthologs of the named species (Ehians  Git3  nav) "
	print "default values: file_extension=fna "	
	Print "prints also for each ortholog the number of sequences, the coefficient of variation of the sizes, and the individual sizes" 
	flag_exit=1; exit
	}
	

	 
BEGIN{	if (ARGC==1){usage(dummy)}	}


(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 (outdir !=""){out_dir=outdir}
		if (out_dir != ""){out_dir = out_dir "/"}  #ensures that a "/" is present in the end
		log_output_file=out_dir output_prefix".ortholog_counts"
		}

(NR>=1){
		current_ortholog=$1
		multifasta_ouput_file = out_dir  output_prefix current_ortholog "." file_extension       # Ehians_Git3_nav_EOG09150006.fna or Ehians_Git3_nav_EOG09150006.faa
		total_ortholog_count ++
		shared_ortholog_count=0
		ortholog_count_array[current_ortholog]=0
		flag_first_saved_sequence=0
		for (i=1; i<=n_species; i++){
				current_species=species_array[i]
				ortholog_sizes_array[current_ortholog,current_species] = 0
				current_file=current_species "/" current_ortholog "." file_extension
				#sed_command="sed \47/>/{s/>/>" current_species "_" current_ortholog " /}\47 " current_file   # sed '/>/{s/>/>Ehians_EOG09150006 /}' Ehians/EOG09150006.fna  
				sed_command="sed \47/>/{s/>.*/>" current_species "_" current_ortholog " /}\47 " current_file   # sed '/>/{s/>.*/>Ehians_EOG09150006 /}' Ehians/EOG09150006.fna
				if (system("test -f " current_file)){				#    0 if the expression is true,   1 if the expression is false
						#print "file does not exist: " current_file			
						}
				else{
						if (flag_first_saved_sequence==0){system(sed_command "  > " multifasta_ouput_file )}  #avoids accumulating sequences when running multiple times the program.
						else                             {system(sed_command " >> " multifasta_ouput_file )}
						ortholog_count_array[current_ortholog] ++
						species_array2[current_species] ++
						command_count_char = "tr -d '\n\r' < " current_file " | wc -m"
						command_count_char | getline sequence_size
						ortholog_sizes_array[current_ortholog,current_species] = sequence_size
						close(command_count_char)
						flag_first_saved_sequence ++
						}		
				}
		}

		

		

END{
	if (flag_exit){exit}
	# for (ortho in ortholog_count_array){
			# print ortho , ortholog_count_array[ortho] > output_prefix".ortholog_counts"   #useful to select the genes thta have orthologs in more or all species
			# }
	print "species","found_genes"
    for (i=1; i <= n_species; i++){
		     current_species=species_array[i]
			 print current_species , species_array2[current_species]
	       }
			
	printf("%s\t%s\t%s\t" , "ortholog", "count" , "CV") > log_output_file
    for (i=1; i <= n_species; i++){printf("%s\t",species_array[i]) > log_output_file } 
	print "" > log_output_file
	for (ortho in ortholog_count_array){	
#			printf("%s\t%i\t", ortho, ortholog_count_array[ortho] ) > log_output_file
			avg=0; CV=0; sum=0; sum_squares=0; var=0; n=0
			for(i=1; i<=n_species; i++){
					current_species=species_array[i]	
					size=ortholog_sizes_array[ortho,current_species]
					sum = sum + size
					sum_squares = sum_squares + size^2
					if (size > 0) {n ++} #excludes the missing data from the calculation of the CV (Coef_Variation)
					}
			printf("%s\t%i\t", ortho, n ) > log_output_file
			if (n>1) {avg=sum/n ; var=sum_squares/n - avg^2; CV=100*sqrt(var)/avg
					  printf("%3.1f\t", CV ) > log_output_file
					 }
			else     {printf("%s\t", "." ) > log_output_file}
			for(i=1; i<=n_species; i++){
					current_species=species_array[i]
					printf("%i\t", ortholog_sizes_array[ortho,current_species] ) > log_output_file
					}
			print "" > log_output_file 
			}			
	}


	
	# 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)
     # }'