main (int argc, char ** argv )
{
char *seqfile; /* name of sequence file */
SQINFO sqinfo; /* extra info about sequence */
SQFILE *dbfp; /* open sequence file */
int fmt,ofmt=106; /* format of seqfile */
/* 106 is PHYLIP format in SQUID */
char *seq; /* sequence */
int type; /* kAmino, kDNA, kRNA, or kOtherSeq */
sequence * seqs, * cds_seqs;
sequence tmp_seqs[2], tmp_cds_seqs[2];
char *optname;
char *optarg, *t;
int optind;
int be_quiet;
int seqct = 0,cdsct = 0;
int min_aln_len = 0;
int do_oneline = 0;
char * output_filename = 0, *submat_file = 0;
int showaln = 1;
int showheader=1;
FILE *ofd, *fd;
alignment *cds_aln;
alignment * opt_alignment = NULL; /* place for pairwise alignment */
int len,i,j, k, jk,ik,aln_count, rc;
pairwise_distances pwMLdist, pwNGdist;
int firsttime = 1;
struct timeval tp;
pwMLdist.N = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.dN = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.S = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.dS = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.dNdS = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.SEdS = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.SEdN = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.t = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwMLdist.kappa= make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwNGdist.dN = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwNGdist.dS = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
pwNGdist.dNdS = make_double_matrix(NUM_PW_SEQS,NUM_PW_SEQS);
/*
pwMLdist.N = pwMLdist.dN = pwMLdist.S = 0;
pwMLdist.dS = pwMLdist.dNdS = pwMLdist.SEdS = 0;
pwMLdist.SEdN = pwMLdist.t = pwMLdist.kappa= 0;
pwNGdist.dN = pwNGdist.dS = pwNGdist.dNdS = 0;
*/
Alntype = default_aln_type;
/* Command line Parse */
fmt = SQFILE_UNKNOWN; /* default: autodetect format */
be_quiet = FALSE;
type = kOtherSeq;
/* for our purposes this is only pairwise alignments, but
* would rather do it correctly in case we move to MSA case
*/
while (Getopt(argc, argv, OPTIONS, NOPTIONS, usage,
&optind, &optname, &optarg))
{
if (strcmp(optname, "--matrix") == 0) submat_file = optarg;
else if (strcmp(optname, "--quiet") == 0) be_quiet = TRUE;
else if (strcmp(optname, "--gapopen") == 0) {
Gapopen = atoi(optarg);
if( Gapopen < 0 ) Gapopen *= -1;
} else if (strcmp(optname, "--gapext") == 0) {
Gapext = atoi(optarg);
if( Gapext < 0 ) Gapext *= -1;
} else if (strcmp(optname, "--informat") == 0) {
fmt = String2SeqfileFormat(optarg);
if (fmt == SQFILE_UNKNOWN)
Die("unrecognized sequence file format \"%s\"", optarg);
} else if (strcmp(optname, "--outformat") == 0) {
ofmt = String2SeqfileFormat(optarg);
if (ofmt == SQFILE_UNKNOWN)
Die("unrecognized sequence file format \"%s\"", optarg);
} else if( strcmp(optname, "--global") == 0 ) {
Alntype = global;
} else if (strcmp(optname, "-h") == 0) {
puts(usage);
puts(experts);
exit(EXIT_SUCCESS);
} else if ( strcmp(optname, "-v") == 0 ) {
Verbose = 1;
} else if ( strcmp(optname, "--gapchar") == 0 ) {
GapChar = optarg[0];
} else if( strcmp(optname, "--output") == 0 ) {
output_filename = optarg;
} else if( strcmp(optname, "--showtable" ) == 0 ) {
showaln = 0;
} else if( strcmp(optname, "--noheader" ) == 0 ) {
showheader = 0;
}
}
if (argc - optind < 1) Die("%s\n", usage);
if( ! submat_file ) {
if( (t = getenv("SUBOPTDIR")) != 0 ||
(t = getenv("SUBOPT_DIR")) != 0 ) {
submat_file = calloc(strlen(t) + 24, sizeof(char));
sprintf(submat_file, "%s/%s",t,Default_submat);
} else {
submat_file = calloc(strlen((void *)Default_submat) + 24, sizeof(char));
sprintf(submat_file, "../%s",Default_submat);
}
}
/* open matrix */
fd = fopen(submat_file, "r");
if( ! ParsePAMFile(fd,&ScoringMatrix, &MatrixScale) ) {
fprintf(stderr, "Cannot parse or open matrix file %s\n",submat_file);
free(submat_file);
exit(EXIT_SUCCESS);
}
if( output_filename && strlen(output_filename) != 1 &&
output_filename[0] != '-') {
ofd = fopen(output_filename,"w");
if( ! ofd ) {
fprintf(stderr, "could not open file %s",output_filename);
goto end;
}
} else
ofd = stdout;
while( optind < argc ) {
seqfile = argv[optind++];
/* Try to work around inability to autodetect from a pipe or .gz:
* assume FASTA format
*/
if (fmt == SQFILE_UNKNOWN &&
(Strparse("^.*\\.gz$", seqfile, 0) || strcmp(seqfile, "-") == 0))
fmt = SQFILE_FASTA;
if ((dbfp = SeqfileOpen(seqfile, fmt, NULL)) == NULL)
Die("Failed to open sequence file %s for reading", seqfile);
while (ReadSeq(dbfp, dbfp->format, &seq, &sqinfo))
{
FreeSequence(NULL, &sqinfo);
seqct++;
}
cds_seqs = (sequence *)calloc(seqct, sizeof(sequence));
seqs = (sequence *)calloc(seqct, sizeof(sequence));
SeqfileRewind(dbfp);
seqct=0;
while (ReadSeq(dbfp, dbfp->format, &seq, &sqinfo))
{
sqinfo.type = Seqtype(seq);
if( sqinfo.type == kDNA || sqinfo.type == kRNA ) {
seqs[seqct].seqstr = Translate(seq,stdcode1);
/* Let's remove the last codon if it is a stop codon */
len = strlen(seqs[seqct].seqstr);
if( Verbose )
fprintf(stderr,"seqct is %d length is %d\n",seqct,
len);
if( seqs[seqct].seqstr[len-1] == '*' ) {
seqs[seqct].seqstr[len-1] = '\0';
seq[strlen(seq) - 3] = '\0';
}
cds_seqs[cdsct].seqstr = seq;
seqs[seqct].seqname = calloc(strlen(sqinfo.name)+1,sizeof(char));
cds_seqs[cdsct].seqname = calloc(strlen(sqinfo.name)+1,sizeof(char));
strcpy(seqs[seqct].seqname,sqinfo.name );
strcpy(cds_seqs[cdsct].seqname,sqinfo.name);
cds_seqs[cdsct].length = sqinfo.len;
cds_seqs[cdsct].alphabet = ( sqinfo.type == kDNA ) ? dna : rna;
seqs[seqct].length = strlen(seqs[seqct].seqstr);
seqs[seqct].alphabet = protein;
cdsct++; seqct++;
} else {
fprintf(stderr,"Expect CDS sequences (DNA or RNA) not Protein\n");
goto end;
}
FreeSequence(NULL, &sqinfo);
if( Verbose && seqct > 3 )
break;
}
if( seqct < 2 ) {
fprintf(stderr,"Must have provided a valid file with at least 2 sequences in it");
goto end;
}
for( i=0; i < seqct; i++ ) {
for(k=i+1; k < seqct; k++ ) {
if( (opt_alignment = (alignment *)calloc(1,sizeof(alignment *))) == NULL) {
fprintf(stderr,"Could not allocate memory\n");
goto end;
}
opt_alignment->msa = NULL;
rc = optimal_align(&seqs[i],&seqs[k],opt_alignment);
if( rc != 1 ) {
fprintf(stderr,"Could not make an optimal alignment\n");
goto end;
} else {
tmp_cds_seqs[0] = cds_seqs[i];
tmp_cds_seqs[1] = cds_seqs[k];
rc = mrtrans(opt_alignment, tmp_cds_seqs, &cds_aln,0);
if( rc != 0 ) {
fprintf(stderr, "Could not map the coding sequence to the protein alignemnt for aln %d: %d\n",i,rc);
goto end;
}
if( showaln ) {
if( ofmt >= 100 ) {
MSAFileWrite(ofd,cds_aln->msa, ofmt,do_oneline);
} else {
for(j=0; j < cds_aln->msa->nseq; j++ ) {
WriteSeq(ofd, ofmt,
cds_aln->msa->aseq[j],
&(cds_aln->sqinfo[j]) );
}
}
} else {
if( showheader && firsttime ) {
fprintf(ofd,"SEQ1\tSEQ2\tSCORE\tdN\tdS\tOMEGA\tN\tS\tkappa\tt\tLENGTH\n");
firsttime = 0;
}
if( do_kaks_yn00(cds_aln->msa, &pwMLdist,&pwNGdist) < 0 ) {
fprintf(stderr, "warning: problem with align for %s %s\n",
cds_aln->msa->sqname[0], cds_aln->msa->sqname[1]);
continue;
}
for(ik = 0; ik < NUM_PW_SEQS; ik++ ) {
for( jk = ik+1; jk < NUM_PW_SEQS; jk++ ) {
fprintf(ofd,"%s\t%s\t%d\t%f\t%f\t%f\t%f\t%f\t%f\t%f\t%d\n",
cds_aln->sqinfo[ik].name,
cds_aln->sqinfo[jk].name,
opt_alignment->score,
pwMLdist.dN[ik][jk],pwMLdist.dS[ik][jk],
pwMLdist.dNdS[ik][jk],
pwMLdist.N[ik][jk],
pwMLdist.S[ik][jk],
pwMLdist.kappa[ik][jk],
pwMLdist.t[ik][jk],
opt_alignment->msa->alen);
}
}
}
}
cleanup_alignment(cds_aln);
cleanup_alignment(opt_alignment);
}
}
}
if( ofd && ofd != stdout )
fclose(ofd);
end:
free(submat_file);
Free2DArray((void **)ScoringMatrix,27);
for(i =0; i< seqct; i++ ) {
free(seqs[i].seqstr);
free(seqs[i].seqname);
seqs[i].seqstr = seqs[i].seqname = 0;
}
for(i = 0; i < cdsct; i++) {
free(cds_seqs[i].seqstr);
free(cds_seqs[i].seqname);
cds_seqs[i].seqstr = cds_seqs[i].seqname = 0;
}
cleanup_matrix((void **)pwMLdist.N,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.dN,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.S,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.dS,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.SEdS,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.SEdN,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.t,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.dNdS,NUM_PW_SEQS);
cleanup_matrix((void **)pwMLdist.kappa,NUM_PW_SEQS);
cleanup_matrix((void **)pwNGdist.dN,NUM_PW_SEQS);
cleanup_matrix((void **)pwNGdist.dS,NUM_PW_SEQS);
cleanup_matrix((void **)pwNGdist.dNdS,NUM_PW_SEQS);
free(pwNGdist.dNdS);
free(pwNGdist.dN);
free(pwNGdist.dS);
free(pwMLdist.dNdS);
free(pwMLdist.dN);
free(pwMLdist.dS);
free(pwMLdist.N);
free(pwMLdist.S);
free(pwMLdist.SEdS);
free(pwMLdist.SEdN);
free(pwMLdist.t);
free(pwMLdist.kappa);
return 0;
}
/**
* @brief Parse command line parameters. Will exit if help/usage etc
* are called or or call Log(&rLog, LOG_FATAL, ) if an error was detected.
*
* @param[out] user_opts
* User parameter struct, with defaults already set.
* @param[in] argc
* mains argc
* @param[in] argv
* mains argv
*
*/
void
ParseCommandLine(cmdline_opts_t *user_opts, int argc, char **argv)
{
/* argtable command line parsing:
* see
* http://argtable.sourceforge.net/doc/argtable2-intro.html
*
* basic structure is: arg_xxxN:
* xxx can be int, lit, db1, str, rex, file or date
* If N = 0, arguments may appear zero-or-once; N = 1 means
* exactly once, N = n means up to maxcount times
*
*
* @note: changes here, might also affect main.cpp:ConvertOldCmdLine()
*
*/
struct arg_rem *rem_seq_input = arg_rem(NULL, "\nSequence Input:");
struct arg_file *opt_seqin = arg_file0("i", "in,infile",
"{<file>,-}",
"Multiple sequence input file (- for stdin)");
struct arg_file *opt_hmm_in = arg_filen(NULL, "hmm-in", "<file>",
/*min*/ 0, /*max*/ 128,
"HMM input files");
struct arg_lit *opt_dealign = arg_lit0(NULL, "dealign",
"Dealign input sequences");
struct arg_file *opt_profile1 = arg_file0(NULL, "profile1,p1",
"<file>",
"Pre-aligned multiple sequence file (aligned columns will be kept fix)");
struct arg_file *opt_profile2 = arg_file0(NULL, "profile2,p2",
"<file>",
"Pre-aligned multiple sequence file (aligned columns will be kept fix)");
struct arg_str *opt_seqtype = arg_str0("t", "seqtype",
"{Protein, RNA, DNA}",
"Force a sequence type (default: auto)");
/* struct arg_lit *opt_force_protein = arg_lit0(NULL, "protein",
"Set sequence type to protein even if Clustal guessed nucleic acid"); */
struct arg_str *opt_infmt = arg_str0(NULL, "infmt",
"{a2m=fa[sta],clu[stal],msf,phy[lip],selex,st[ockholm],vie[nna]}",
"Forced sequence input file format (default: auto)");
struct arg_rem *rem_guidetree = arg_rem(NULL, "\nClustering:");
struct arg_str *opt_pairdist = arg_str0("p", "pairdist",
"{ktuple}",
"Pairwise alignment distance measure");
struct arg_file *opt_distmat_in = arg_file0(NULL, "distmat-in",
"<file>",
"Pairwise distance matrix input file (skips distance computation)");
struct arg_file *opt_distmat_out = arg_file0(NULL, "distmat-out",
"<file>",
"Pairwise distance matrix output file");
struct arg_file *opt_guidetree_in = arg_file0(NULL, "guidetree-in",
"<file>",
"Guide tree input file (skips distance computation and guide-tree clustering step)");
struct arg_file *opt_guidetree_out = arg_file0(NULL, "guidetree-out",
"<file>",
"Guide tree output file");
/* AW: mbed is default since at least R253
struct arg_lit *opt_mbed = arg_lit0(NULL, "mbed",
"Fast, Mbed-like clustering for guide-tree calculation");
struct arg_lit *opt_mbed_iter = arg_lit0(NULL, "mbed-iter",
"Use Mbed-like clustering also during iteration");
*/
/* Note: might be better to use arg_str (mbed=YES/NO) but I don't want to introduce an '=' into pipeline, FS, r250 -> */
struct arg_lit *opt_full = arg_lit0(NULL, "full",
"Use full distance matrix for guide-tree calculation (might be slow; mBed is default)");
struct arg_lit *opt_full_iter = arg_lit0(NULL, "full-iter",
"Use full distance matrix for guide-tree calculation during iteration (might be slowish; mBed is default)");
struct arg_str *opt_clustering = arg_str0("c", "clustering",
"{UPGMA}",
"Clustering method for guide tree");
struct arg_rem *rem_aln_output = arg_rem(NULL, "\nAlignment Output:");
struct arg_file *opt_outfile = arg_file0("o", "out,outfile",
"{file,-}",
"Multiple sequence alignment output file (default: stdout)");
struct arg_str *opt_outfmt = arg_str0(NULL, "outfmt",
"{a2m=fa[sta],clu[stal],msf,phy[lip],selex,st[ockholm],vie[nna]}",
"MSA output file format (default: fasta)");
struct arg_rem *rem_iteration = arg_rem(NULL, "\nIteration:");
struct arg_str *opt_num_iterations = arg_str0(NULL, "iterations,iter",
/* FIXME "{<n>,auto}", "Number of combined guide-tree/HMM iterations"); */
"<n>", "Number of (combined guide-tree/HMM) iterations");
struct arg_int *opt_max_guidetree_iterations = arg_int0(NULL, "max-guidetree-iterations",
"<n>", "Maximum number guidetree iterations");
struct arg_int *opt_max_hmm_iterations = arg_int0(NULL, "max-hmm-iterations",
"<n>", "Maximum number of HMM iterations");
struct arg_rem *rem_limits = arg_rem(NULL, "\nLimits (will exit early, if exceeded):");
struct arg_int *opt_max_seq = arg_int0(NULL, "maxnumseq", "<n>",
"Maximum allowed number of sequences");
struct arg_int *opt_max_seqlen = arg_int0(NULL, "maxseqlen", "<l>",
"Maximum allowed sequence length");
struct arg_rem *rem_misc = arg_rem(NULL, "\nMiscellaneous:");
struct arg_lit *opt_autooptions = arg_lit0(NULL, "auto",
"Set options automatically (might overwrite some of your options)");
struct arg_int *opt_threads = arg_int0(NULL, "threads", "<n>",
"Number of processors to use");
struct arg_file *opt_logfile = arg_file0("l", "log",
"<file>",
"Log all non-essential output to this file");
struct arg_lit *opt_help = arg_lit0("h", "help",
"Print this help and exit");
struct arg_lit *opt_version = arg_lit0(NULL, "version",
"Print version information and exit");
struct arg_lit *opt_long_version = arg_lit0(NULL, "long-version",
"Print long version information and exit");
struct arg_lit *opt_verbose = arg_litn("v", "verbose",
0, 3,
"Verbose output (increases if given multiple times)");
struct arg_lit *opt_force = arg_lit0(NULL, "force",
"Force file overwriting");
struct arg_int *opt_macram = arg_int0(NULL, "MAC-RAM", "<n>", /* keep this quiet for the moment, FS r240 -> */
NULL/*"maximum amount of RAM to use for MAC algorithm (in MB)"*/);
struct arg_end *opt_end = arg_end(10); /* maximum number of errors
* to store */
void *argtable[] = {rem_seq_input,
opt_seqin,
opt_hmm_in,
opt_dealign,
opt_profile1,
opt_profile2,
opt_seqtype,
/* opt_force_protein, */
opt_infmt,
rem_guidetree,
#if 0
/* no other options then default available or not implemented */
opt_pairdist,
#endif
opt_distmat_in,
opt_distmat_out,
opt_guidetree_in,
opt_guidetree_out,
opt_full, /* FS, r250 -> */
opt_full_iter, /* FS, r250 -> */
#if 0
/* no other options then default available */
opt_clustering,
#endif
rem_aln_output,
opt_outfile,
opt_outfmt,
rem_iteration,
opt_num_iterations,
opt_max_guidetree_iterations,
opt_max_hmm_iterations,
rem_limits,
opt_max_seq,
opt_max_seqlen,
rem_misc,
opt_autooptions,
opt_threads,
opt_logfile,
opt_help,
opt_verbose,
opt_version,
opt_long_version,
opt_force,
opt_macram, /* FS, r240 -> r241 */
opt_end};
int nerrors;
/* Verify the argtable[] entries were allocated sucessfully
*/
if (arg_nullcheck(argtable)) {
Log(&rLog, LOG_FATAL, "insufficient memory (for argtable allocation)");
}
/* Parse the command line as defined by argtable[]
*/
nerrors = arg_parse(argc, argv, argtable);
/* Special case: '--help' takes precedence over error reporting
*/
if (opt_help->count > 0) {
printf("%s - %s (%s)\n", PACKAGE_NAME, PACKAGE_VERSION, PACKAGE_CODENAME);
printf("\n");
printf("If you like Clustal-Omega please cite:\n%s\n", CITATION);
printf("If you don't like Clustal-Omega, please let us know why (and cite us anyway).\n");
/* printf("You can contact reach us under %s\n", PACKAGE_BUGREPORT); */
printf("\n");
printf("Check http://www.clustal.org for more information and updates.\n");
printf("\n");
printf("Usage: %s", basename(argv[0]));
arg_print_syntax(stdout,argtable, "\n");
printf("\n");
printf("A typical invocation would be: %s -i my-in-seqs.fa -o my-out-seqs.fa -v\n",
basename(argv[0]));
printf("See below for a list of all options.\n");
arg_print_glossary(stdout, argtable, " %-25s %s\n");
arg_freetable(argtable, sizeof(argtable)/sizeof(argtable[0]));
exit(EXIT_SUCCESS);
}
/* Special case: '--version' takes precedence over error reporting
*/
if (opt_version->count > 0) {
printf("%s\n", PACKAGE_VERSION);
arg_freetable(argtable,sizeof(argtable)/sizeof(argtable[0]));
exit(EXIT_SUCCESS);
}
/* Special case: '--long-version' takes precedence over error reporting
*/
if (opt_long_version->count > 0) {
char zcLongVersion[1024];
PrintLongVersion(zcLongVersion, sizeof(zcLongVersion));
printf("%s\n", zcLongVersion);
arg_freetable(argtable,sizeof(argtable)/sizeof(argtable[0]));
exit(EXIT_SUCCESS);
}
/* If the parser returned any errors then display them and exit
*/
if (nerrors > 0) {
/* Display the error details contained in the arg_end struct.*/
arg_print_errors(stdout, opt_end, PACKAGE);
fprintf(stderr, "For more information try: %s --help\n", argv[0]);
arg_freetable(argtable,sizeof(argtable)/sizeof(argtable[0]));
exit(EXIT_FAILURE);
}
/* ------------------------------------------------------------
*
* Command line successfully parsed. Now transfer values to
* user_opts. While doing so, make sure that given input files
* exist and given output files are writable do not exist, or if
* they do, should be overwritten.
*
* No logic checks here! They are done in a different function
*
* ------------------------------------------------------------*/
/* not part of user_opts because it declared in src/util.h */
if (0 == opt_verbose->count) {
rLog.iLogLevelEnabled = LOG_WARN;
} else if (1 == opt_verbose->count) {
rLog.iLogLevelEnabled = LOG_INFO;
} else if (2 == opt_verbose->count) {
rLog.iLogLevelEnabled = LOG_VERBOSE;
} else if (3 == opt_verbose->count) {
rLog.iLogLevelEnabled = LOG_DEBUG;
}
user_opts->aln_opts.bAutoOptions = opt_autooptions->count;
user_opts->bDealignInputSeqs = opt_dealign->count;
/* NOTE: full distance matrix used to be default - there was
--mbed flag but no --full flag after r250 decided that mBed
should be default - now need --full flag to turn off mBed.
wanted to retain mBed Boolean, so simply added --full flag. if
both flags set (erroneously) want --mbed to overwrite --full,
therefore do --full 1st, the --mbed, FS, r250 */
if (opt_full->count){
user_opts->aln_opts.bUseMbed = FALSE;
}
if (opt_full_iter->count){
user_opts->aln_opts.bUseMbedForIteration = FALSE;
}
user_opts->bForceFileOverwrite = opt_force->count;
/* log-file
*/
if (opt_logfile->count > 0) {
user_opts->pcLogFile = CkStrdup(opt_logfile->filename[0]);
/* warn if already exists or not writable */
if (FileExists(user_opts->pcLogFile) && ! user_opts->bForceFileOverwrite) {
Log(&rLog, LOG_FATAL, "%s '%s'. %s",
"Cowardly refusing to overwrite already existing file",
user_opts->pcLogFile,
"Use --force to force overwriting.");
}
if (! FileIsWritable(user_opts->pcLogFile)) {
Log(&rLog, LOG_FATAL, "Sorry, I do not have permission to write to file '%s'.",
user_opts->pcLogFile);
}
}
/* normal sequence input (no profile)
*/
if (opt_seqin->count > 0) {
user_opts->pcSeqInfile = CkStrdup(opt_seqin->filename[0]);
}
/* Input limitations
*/
/* maximum number of sequences */
if (opt_max_seq->count > 0) {
user_opts->iMaxNumSeq = opt_max_seq->ival[0];
}
/* maximum sequence length */
if (opt_max_seqlen->count > 0) {
user_opts->iMaxSeqLen = opt_max_seqlen->ival[0];
}
/* Output format
*/
if (opt_infmt->count > 0) {
/* avoid gcc warning about discarded qualifier */
char *tmp = (char *)opt_infmt->sval[0];
user_opts->iSeqInFormat = String2SeqfileFormat(tmp);
} else {
user_opts->iSeqInFormat = SQFILE_UNKNOWN;
}
/* Sequence type
*/
if (opt_seqtype->count > 0) {
if (STR_NC_EQ(opt_seqtype->sval[0], "protein")) {
user_opts->iSeqType = SEQTYPE_PROTEIN;
} else if (STR_NC_EQ(opt_seqtype->sval[0], "rna")) {
user_opts->iSeqType = SEQTYPE_RNA;
} else if (STR_NC_EQ(opt_seqtype->sval[0], "dna")) {
user_opts->iSeqType = SEQTYPE_DNA;
} else {
Log(&rLog, LOG_FATAL, "Unknown sequence type '%s'", opt_seqtype->sval[0]);
}
}
/* if (opt_force_protein->count > 0) {
user_opts->iSeqType = SEQTYPE_PROTEIN;
} */
/* Profile input
*/
if (opt_profile1->count > 0) {
user_opts->pcProfile1Infile = CkStrdup(opt_profile1->filename[0]);
if (! FileExists(user_opts->pcProfile1Infile)) {
Log(&rLog, LOG_FATAL, "File '%s' does not exist.", user_opts->pcProfile1Infile);
}
}
if (opt_profile2->count > 0) {
user_opts->pcProfile2Infile = CkStrdup(opt_profile2->filename[0]);
if (! FileExists(user_opts->pcProfile2Infile)) {
Log(&rLog, LOG_FATAL, "File '%s' does not exist.", user_opts->pcProfile2Infile);
}
}
/* HMM input
*/
user_opts->aln_opts.iHMMInputFiles = 0;
user_opts->aln_opts.ppcHMMInput = NULL;
if (opt_hmm_in->count>0) {
int iAux;
user_opts->aln_opts.iHMMInputFiles = opt_hmm_in->count;
user_opts->aln_opts.ppcHMMInput = (char **) CKMALLOC(
user_opts->aln_opts.iHMMInputFiles * sizeof(char*));
for (iAux=0; iAux<opt_hmm_in->count; iAux++) {
user_opts->aln_opts.ppcHMMInput[iAux] = CkStrdup(opt_hmm_in->filename[iAux]);
if (! FileExists(user_opts->aln_opts.ppcHMMInput[iAux])) {
Log(&rLog, LOG_FATAL, "File '%s' does not exist.", user_opts->aln_opts.ppcHMMInput[iAux]);
}
}
}
/* Pair distance method
*/
if (opt_pairdist->count > 0) {
if (STR_NC_EQ(opt_pairdist->sval[0], "ktuple")) {
user_opts->aln_opts.iPairDistType = PAIRDIST_KTUPLE;
} else {
Log(&rLog, LOG_FATAL, "Unknown pairdist method '%s'", opt_pairdist->sval[0]);
}
}
/* Distance matrix input
*/
if (opt_distmat_in->count > 0) {
user_opts->aln_opts.pcDistmatInfile = CkStrdup(opt_distmat_in->filename[0]);
if (! FileExists(user_opts->aln_opts.pcDistmatInfile)) {
Log(&rLog, LOG_FATAL, "File '%s' does not exist.", user_opts->aln_opts.pcDistmatInfile);
}
}
/* Distance matrix output
*/
if (opt_distmat_out->count > 0) {
user_opts->aln_opts.pcDistmatOutfile = CkStrdup(opt_distmat_out->filename[0]);
/* warn if already exists or not writable */
if (FileExists(user_opts->aln_opts.pcDistmatOutfile) && ! user_opts->bForceFileOverwrite) {
Log(&rLog, LOG_FATAL, "%s '%s'. %s",
"Cowardly refusing to overwrite already existing file",
user_opts->aln_opts.pcDistmatOutfile,
"Use --force to force overwriting.");
}
if (! FileIsWritable(user_opts->aln_opts.pcDistmatOutfile)) {
Log(&rLog, LOG_FATAL, "Sorry, I do not have permission to write to file '%s'.",
user_opts->aln_opts.pcDistmatOutfile);
}
}
/* Clustering
*
*/
if (opt_clustering->count > 0) {
if (STR_NC_EQ(opt_clustering->sval[0], "upgma")) {
user_opts->aln_opts.iClusteringType = CLUSTERING_UPGMA;
} else {
Log(&rLog, LOG_FATAL, "Unknown guide-tree clustering method '%s'", opt_clustering->sval[0]);
}
}
/* Guidetree input
*/
if (opt_guidetree_in->count > 0) {
user_opts->aln_opts.pcGuidetreeInfile = CkStrdup(opt_guidetree_in->filename[0]);
if (! FileExists(user_opts->aln_opts.pcGuidetreeInfile)) {
Log(&rLog, LOG_FATAL, "File '%s' does not exist.", user_opts->aln_opts.pcGuidetreeInfile);
}
}
/* Guidetree output
*/
if (opt_guidetree_out->count > 0) {
user_opts->aln_opts.pcGuidetreeOutfile = CkStrdup(opt_guidetree_out->filename[0]);
/* warn if already exists or not writable */
if (FileExists(user_opts->aln_opts.pcGuidetreeOutfile) && ! user_opts->bForceFileOverwrite) {
Log(&rLog, LOG_FATAL, "%s '%s'. %s",
"Cowardly refusing to overwrite already existing file",
user_opts->aln_opts.pcGuidetreeOutfile,
"Use --force to force overwriting.");
}
if (! FileIsWritable(user_opts->aln_opts.pcGuidetreeOutfile)) {
Log(&rLog, LOG_FATAL, "Sorry, I do not have permission to write to file '%s'.",
user_opts->aln_opts.pcGuidetreeOutfile);
}
}
/* max guidetree iterations
*/
if (opt_max_guidetree_iterations->count > 0) {
user_opts->aln_opts.iMaxGuidetreeIterations = opt_max_guidetree_iterations->ival[0];
}
/* max guidetree iterations
*/
if (opt_max_hmm_iterations->count > 0) {
user_opts->aln_opts.iMaxHMMIterations = opt_max_hmm_iterations->ival[0];
}
/* number of iterations
*/
if (opt_num_iterations->count > 0) {
if (STR_NC_EQ(opt_num_iterations->sval[0], "auto")) {
Log(&rLog, LOG_FATAL, "Automatic iteration not supported at the moment.");
user_opts->aln_opts.bIterationsAuto = TRUE;
} else {
int iAux;
user_opts->aln_opts.bIterationsAuto = FALSE;
for (iAux=0; iAux<(int)strlen(opt_num_iterations->sval[0]); iAux++) {
if (! isdigit(opt_num_iterations->sval[0][iAux])) {
Log(&rLog, LOG_FATAL, "Couldn't iteration parameter: %s",
opt_num_iterations->sval[0]);
}
}
user_opts->aln_opts.iNumIterations = atoi(opt_num_iterations->sval[0]);
}
}
/* Alignment output
*/
if (opt_outfile->count > 0) {
user_opts->pcAlnOutfile = CkStrdup(opt_outfile->filename[0]);
/* warn if already exists or not writable */
if (FileExists(user_opts->pcAlnOutfile) && ! user_opts->bForceFileOverwrite) {
Log(&rLog, LOG_FATAL, "%s '%s'. %s",
"Cowardly refusing to overwrite already existing file",
user_opts->pcAlnOutfile,
"Use --force to force overwriting.");
}
if (! FileIsWritable(user_opts->pcAlnOutfile)) {
Log(&rLog, LOG_FATAL, "Sorry, I do not have permission to write to file '%s'.",
user_opts->pcAlnOutfile);
}
}
/* Output format
*/
if (opt_outfmt->count > 0) {
/* avoid gcc warning about discarded qualifier */
char *tmp = (char *)opt_outfmt->sval[0];
user_opts->iAlnOutFormat = String2SeqfileFormat(tmp);
if (SQFILE_UNKNOWN == user_opts->iAlnOutFormat) {
Log(&rLog, LOG_FATAL, "Unknown output format '%s'", opt_outfmt->sval[0]);
}
}
/* Number of threads
*/
#ifdef HAVE_OPENMP
if (opt_threads->count > 0) {
if (opt_threads->ival[0] <= 0) {
Log(&rLog, LOG_FATAL, "Changing number of threads to %d doesn't make sense.",
opt_threads->ival[0]);
}
user_opts->iThreads = opt_threads->ival[0];
}
#else
if (opt_threads->count > 0) {
if (opt_threads->ival[0] > 1) {
Log(&rLog, LOG_FATAL, "Cannot change number of threads to %d. %s was build without OpenMP support.",
opt_threads->ival[0], PACKAGE_NAME);
}
}
#endif
/* max MAC RAM (maximum amount of RAM set aside for MAC algorithm)
*/
if (opt_macram->count > 0) { /* FS, r240 -> r241 */
user_opts->aln_opts.rHhalignPara.iMacRamMB = opt_macram->ival[0];
}
arg_freetable(argtable,sizeof(argtable)/sizeof(argtable[0]));
UserOptsLogicCheck(user_opts);
return;
}
