void inputancestors(boolean *anczero0, boolean *ancone0)
{
/* reads the ancestral states for each character */
/* used in dollop, dolmove, dolpenny, mix, move, & penny */
long i;
Char ch;
for (i = 0; i < (chars); i++) {
anczero0[i] = true;
ancone0[i] = true;
do {
if (eoln(ancfile))
scan_eoln(ancfile);
ch = gettc(ancfile);
if (ch == '\n')
ch = ' ';
} while (ch == ' ');
if (ch == 'p')
ch = 'P';
if (ch == 'b')
ch = 'B';
if (strchr("10PB?",ch) != NULL){
anczero0[i] = (ch == '1') ? false : anczero0[i];
ancone0[i] = (ch == '0') ? false : ancone0[i];
} else {
printf("BAD ANCESTOR STATE: %cAT CHARACTER %4ld\n", ch, i + 1);
exxit(-1);
}
}
scan_eoln(ancfile);
} /* inputancestorsnew */
void seqboot_inputnumbers()
{
/* read numbers of species and of sites */
long i;
fscanf(infile, "%ld%ld", &spp, &sites);
loci = sites;
maxalleles = 1;
if (data == restsites && enzymes)
fscanf(infile, "%ld", &nenzymes);
if (data == genefreqs) {
alleles = (long *)Malloc(sites*sizeof(long));
scan_eoln(infile);
sites = 0;
for (i = 0; i < (loci); i++) {
if (eoln(infile))
scan_eoln(infile);
fscanf(infile, "%ld", &alleles[i]);
if (alleles[i] > maxalleles)
maxalleles = alleles[i];
if (all)
sites += alleles[i];
else
sites += alleles[i] - 1;
}
if (!all)
maxalleles--;
scan_eoln(infile);
}
} /* seqboot_inputnumbers */
void getalleles()
{ /* set up number of alleles at loci */
long i, j, m;
if (!firstset)
samenumsp(&loci, ith);
if (contchars ) {
totalleles = loci;
for (i = 1; i <= loci; i++) {
locus[i - 1] = i;
alleles[i - 1] = 1;
}
df = loci;
} else {
totalleles = 0;
scan_eoln(infile);
if (printdata) {
fprintf(outfile, "\nNumbers of alleles at the loci:\n");
fprintf(outfile, "------- -- ------- -- --- -----\n\n");
}
for (i = 1; i <= loci; i++) {
if (eoln(infile))
scan_eoln(infile);
if (fscanf(infile, "%ld", &alleles[i - 1]) != 1) {
printf("ERROR: Unable to read number of alleles at locus %ld\n", i);
exxit(-1);
}
if (alleles[i - 1] <= 0) {
printf("ERROR: Bad number of alleles: %ld at locus %ld\n", alleles[i-1], i);
exxit(-1);
}
totalleles += alleles[i - 1];
if (printdata)
fprintf(outfile, "%4ld", alleles[i - 1]);
}
locus = (long *)Malloc(totalleles*sizeof(long));
m = 0;
for (i = 1; i <= loci; i++) {
for (j = 0; j < alleles[i - 1]; j++)
locus[m+j] = i;
m += alleles[i - 1];
}
df = totalleles - loci;
}
allocview(&curtree, nonodes2, totalleles);
if (!usertree) {
allocview(&bestree, nonodes2, totalleles);
allocview(&priortree, nonodes2, totalleles);
if (njumble > 1)
allocview(&bestree2, nonodes2, totalleles);
}
for (i = 0; i < spp; i++)
x[i] = (phenotype3)Malloc(totalleles*sizeof(double));
pbar = (double *)Malloc(totalleles*sizeof(double));
if (usertree)
for (i = 0; i < MAXSHIMOTREES; i++)
l0gf[i] = (double *)Malloc(totalleles*sizeof(double));
if (printdata)
putc('\n', outfile);
} /* getalleles */
void inputoptions()
{
/* read the options information */
Char ch;
long i, extranum, cursp, curst;
if (!firstset) {
if (eoln(infile))
scan_eoln(infile);
fscanf(infile, "%ld%ld", &cursp, &curst);
if (cursp != spp) {
printf("\nERROR: INCONSISTENT NUMBER OF SPECIES IN DATA SET %4ld\n",
ith);
exxit(-1);
}
sites = curst;
reallocsites();
}
for (i = 1; i <= sites; i++)
weight[i] = 1;
weightsum = sites;
extranum = 0;
fscanf(infile, "%*[ 0-9]");
readoptions(&extranum, "W");
for (i = 1; i <= extranum; i++) {
matchoptions(&ch, "W");
inputweights2(1, sites+1, &weightsum, weight, &weights, "RESTDIST");
}
} /* inputoptions */
void inputcategories()
{
/* reads the categories for each site */
long i;
Phylip_Char ch;
for (i = 1; i < nmlngth; i++)
gettc(infile);
for (i = 0; i < sites; i++) {
do {
if (eoln(infile))
scan_eoln(infile);
ch = gettc(infile);
} while (ch == ' ');
category[i] = ch - '0';
}
scan_eoln(infile);
} /* inputcategories */
void seqboot_inputfactors()
{
long i, j;
Char ch, prevch;
prevch = ' ';
j = 0;
for (i = 0; i < (sites); i++) {
do {
if (eoln(factfile))
scan_eoln(factfile);
ch = gettc(factfile);
} while (ch == ' ');
if (ch != prevch)
j++;
prevch = ch;
factorr[i] = j;
}
scan_eoln(factfile);
} /* seqboot_inputfactors */
void seqboot_inputaux(steptr dataptr, FILE* auxfile)
{ /* input auxiliary option data (mixtures, ancestors, ect) for
new style input, assumes that data is correctly formated
in input files*/
long i, j, k;
Char ch;
j = 0;
k = 1;
for (i = 0; i < (sites); i++) {
do {
if (eoln(auxfile))
scan_eoln(auxfile);
ch = gettc(auxfile);
if (ch == '\n')
ch = ' ';
} while (ch == ' ');
dataptr[i] = ch;
}
scan_eoln(auxfile);
} /* seqboot_inputaux */
void inputmixture(bitptr wagner0)
{
/* input mixture of methods */
/* used in mix, move, & penny */
long i, j, k;
Char ch;
boolean wag;
for (i = 0; i < (words); i++)
wagner0[i] = 0;
j = 0;
k = 1;
for (i = 1; i <= (chars); i++) {
do {
if (eoln(mixfile))
scan_eoln(mixfile);
ch = gettc(mixfile);
if (ch == '\n')
ch = ' ';
} while (ch == ' ');
uppercase(&ch);
wag = false;
if (ch == 'W' || ch == '?')
wag = true;
else if (ch == 'S' || ch == 'C')
wag = false;
else {
printf("BAD METHOD: %c\n", ch);
exxit(-1);
}
if (wag)
wagner0[k - 1] = (long)wagner0[k - 1] | (1L << j);
j++;
if (j > bits) {
j = 1;
k++;
}
}
scan_eoln(mixfile);
} /* inputmixture */
int main(int argc, Char *argv[])
{
int i;
#ifdef MAC
argc = 1; /* macsetup("Fitch",""); */
argv[0]="Fitch";
#endif
init(argc,argv);
progname = argv[0];
openfile(&infile,INFILE,"input file","r",argv[0],infilename);
openfile(&outfile,OUTFILE,"output file","w",argv[0],outfilename);
ibmpc = IBMCRT;
ansi = ANSICRT;
mulsets = false;
datasets = 1;
firstset = true;
doinit();
if (trout)
openfile(&outtree,OUTTREE,"output tree file","w",argv[0],outtreename);
for (i=0;i<spp;++i){
enterorder[i]=0;}
for (ith = 1; ith <= datasets; ith++) {
if (datasets > 1) {
fprintf(outfile, "Data set # %ld:\n\n",ith);
if (progress)
printf("\nData set # %ld:\n\n",ith);
}
fitch_getinput();
for (jumb = 1; jumb <= njumble; jumb++)
maketree();
firstset = false;
if (eoln(infile) && (ith < datasets))
scan_eoln(infile);
}
if (trout)
FClose(outtree);
FClose(outfile);
FClose(infile);
#ifdef MAC
fixmacfile(outfilename);
fixmacfile(outtreename);
#endif
printf("Done.\n\n");
#ifdef WIN32
phyRestoreConsoleAttributes();
#endif
return 0;
}
int main(int argc, Char *argv[])
{ /* main program */
#ifdef MAC
argc = 1; /* macsetup("Neighbor",""); */
argv[0] = "Neighbor";
#endif
init(argc, argv);
openfile(&infile,INFILE,"input file", "r",argv[0],infilename);
openfile(&outfile,OUTFILE,"output file", "w",argv[0],outfilename);
ibmpc = IBMCRT;
ansi = ANSICRT;
mulsets = false;
datasets = 1;
doinit();
if (trout)
openfile(&outtree,OUTTREE,"output tree file", "w",argv[0],outtreename);
ith = 1;
while (ith <= datasets) {
if (datasets > 1) {
fprintf(outfile, "Data set # %ld:\n",ith);
if (progress)
printf("Data set # %ld:\n",ith);
}
inputoptions();
maketree();
if (eoln(infile) && (ith < datasets))
scan_eoln(infile);
ith++;
}
FClose(infile);
FClose(outfile);
FClose(outtree);
freerest();
#ifdef MAC
fixmacfile(outfilename);
fixmacfile(outtreename);
#endif
printf("Done.\n\n");
#ifdef WIN32
phyRestoreConsoleAttributes();
#endif
return 0;
}
void inputoptions()
{
/* input the information on the options */
long i;
scan_eoln(infile);
for (i = 0; i < (chars); i++)
weight[i] = 1;
if (ancvar)
inputancestors(anczero0, ancone0);
if (factors)
inputfactors(chars, factor, &factors);
if (weights)
inputweights(chars, weight, &weights);
putchar('\n');
if (weights)
printweights(stdout, 0, chars, weight, "Characters");
if (factors)
printfactors(stdout, chars, factor, "");
for (i = 0; i < (chars); i++) {
if (!ancvar) {
anczero[i] = true;
ancone[i] = false;
} else {
anczero[i] = anczero0[i];
ancone[i] = ancone0[i];
}
}
if (ancvar)
printancestors(stdout, anczero, ancone);
if (!thresh)
threshold = spp;
questions = false;
for (i = 0; i < (chars); i++) {
questions = (questions || (ancone[i] && anczero[i]));
threshwt[i] = threshold * weight[i];
}
} /* inputoptions */
void inputdata()
{ /* read species data */
long i, j, k, l, m, m0, n, p;
double sum;
if (printdata) {
fprintf(outfile, "\nName");
if (contchars)
fprintf(outfile, " Phenotypes\n");
else
fprintf(outfile, " Gene Frequencies\n");
fprintf(outfile, "----");
if (contchars)
fprintf(outfile, " ----------\n");
else
fprintf(outfile, " ---- -----------\n");
putc('\n', outfile);
if (!contchars) {
for (j = 1; j <= nmlngth - 8; j++)
putc(' ', outfile);
fprintf(outfile, "locus:");
p = 1;
for (j = 1; j <= loci; j++) {
if (all)
n = alleles[j - 1];
else
n = alleles[j - 1] - 1;
for (k = 1; k <= n; k++) {
fprintf(outfile, "%10ld", j);
if (p % 6 == 0 && (all || p < df)) {
putc('\n', outfile);
for (l = 1; l <= nmlngth - 2; l++)
putc(' ', outfile);
}
p++;
}
}
fprintf(outfile, "\n\n");
}
}
for (i = 0; i < spp; i++) {
scan_eoln(infile);
initname(i);
if (printdata)
for (j = 0; j < nmlngth; j++)
putc(nayme[i][j], outfile);
m = 1;
p = 1;
for (j = 1; j <= loci; j++) {
m0 = m;
sum = 0.0;
if (contchars)
n = 1;
else if (all)
n = alleles[j - 1];
else
n = alleles[j - 1] - 1;
for (k = 1; k <= n; k++) {
if (eoln(infile))
scan_eoln(infile);
if (fscanf(infile, "%lf", &x[i][m - 1]) != 1) {
printf("ERROR: unable to read allele frequency"
"for species %ld, locus %ld\n", i+1, j);
exxit(-1);
}
sum += x[i][m - 1];
if (!contchars && x[i][m - 1] < 0.0) {
printf("\n\nERROR: locus %ld in species %ld: an allele", j, i+1);
printf(" frequency is negative\n");
exxit(-1);
}
if (printdata) {
fprintf(outfile, "%10.5f", x[i][m - 1]);
if (p % 6 == 0 && (all || p < df)) {
putc('\n', outfile);
for (l = 1; l <= nmlngth; l++)
putc(' ', outfile);
}
}
p++;
m++;
}
if (all && !contchars) {
if (fabs(sum - 1.0) > epsilon2) {
printf(
"\n\nERROR: Locus %ld in species %ld: frequencies do not add up to 1\n",
j, i + 1);
printf("\nFrequencies are:\n");
for (l = m0; l <= m-3; l++)
printf("%f+", x[i][l]);
printf("%f = %f\n\n", x[i][m-2], sum);
exxit(-1);
} else {
for (l = 0; l <= m-2; l++)
x[i][l] /= sum;
}
}
if (!all && !contchars) {
x[i][m-1] = 1.0 - sum;
if (x[i][m-1] < 0.0) {
if (x[i][m-1] > -epsilon2) {
for (l = 0; l <= m-2; l++)
x[i][l] /= sum;
x[i][m-1] = 0.0;
} else {
printf("\n\nERROR: Locus %ld in species %ld: ", j, i + 1);
printf("frequencies add up to more than 1\n");
printf("\nFrequencies are:\n");
for (l = m0-1; l <= m-3; l++)
printf("%f+", x[i][l]);
printf("%f = %f\n\n", x[i][m-2], sum);
exxit(-1);
}
}
m++;
}
}
if (printdata)
putc('\n', outfile);
}
scan_eoln(infile);
if (printdata)
putc('\n', outfile);
} /* inputdata */
void inputdata(pointptr treenode,boolean dollo,boolean printdata,FILE *outfile)
{
/* input the names and character state data for species */
/* used in Dollop, Dolpenny, Dolmove, & Move */
long i, j, l;
char k;
Char charstate;
/* possible states are '0', '1', 'P', 'B', and '?' */
if (printdata)
headings(chars, "Characters", "----------");
for (i = 0; i < (chars); i++)
extras[i] = 0;
for (i = 1; i <= spp; i++) {
initname(i-1);
if (printdata) {
for (j = 0; j < nmlngth; j++)
putc(nayme[i - 1][j], outfile);
fprintf(outfile, " ");
}
for (j = 0; j < (words); j++) {
treenode[i - 1]->stateone[j] = 0;
treenode[i - 1]->statezero[j] = 0;
}
for (j = 1; j <= (chars); j++) {
k = (j - 1) % bits + 1;
l = (j - 1) / bits + 1;
do {
if (eoln(infile))
scan_eoln(infile);
charstate = gettc(infile);
} while (charstate == ' ' || charstate == '\t');
if (charstate == 'b')
charstate = 'B';
if (charstate == 'p')
charstate = 'P';
if (charstate != '0' && charstate != '1' && charstate != '?' &&
charstate != 'P' && charstate != 'B') {
printf("\n\nERROR: Bad character state: %c ",charstate);
printf("at character %ld of species %ld\n\n", j, i);
exxit(-1);
}
if (printdata) {
newline(outfile, j, 55, nmlngth + 3);
putc(charstate, outfile);
if (j % 5 == 0)
putc(' ', outfile);
}
if (charstate == '1')
treenode[i - 1]->stateone[l - 1] =
((long)treenode[i - 1]->stateone[l - 1]) | (1L << k);
if (charstate == '0')
treenode[i - 1]->statezero[l - 1] =
((long)treenode[i - 1]->statezero[l - 1]) | (1L << k);
if (charstate == 'P' || charstate == 'B') {
if (dollo)
extras[j - 1] += weight[j - 1];
else {
treenode[i - 1]->stateone[l - 1] =
((long)treenode[i - 1]->stateone[l - 1]) | (1L << k);
treenode[i - 1]->statezero[l - 1] =
((long)treenode[i - 1]->statezero[l - 1]) | (1L << k);
}
}
}
scan_eoln(infile);
if (printdata)
putc('\n', outfile);
}
if (printdata)
fprintf(outfile, "\n\n");
} /* inputdata */
void inputdata(boolean replicates, boolean printdata, boolean lower,
boolean upper, vector *x, intvector *reps)
{
/* read in distance matrix */
/* used in fitch & neighbor */
long i=0, j=0, k=0, columns=0;
boolean skipit=false, skipother=false;
if (replicates)
columns = 4;
else
columns = 6;
if (printdata) {
fprintf(outfile, "\nName Distances");
if (replicates)
fprintf(outfile, " (replicates)");
fprintf(outfile, "\n---- ---------");
if (replicates)
fprintf(outfile, "-------------");
fprintf(outfile, "\n\n");
}
for (i = 0; i < spp; i++) {
x[i][i] = 0.0;
scan_eoln(infile);
initname(i);
for (j = 0; j < spp; j++) {
skipit = ((lower && j + 1 >= i + 1) || (upper && j + 1 <= i + 1));
skipother = ((lower && i + 1 >= j + 1) || (upper && i + 1 <= j + 1));
if (!skipit) {
if (eoln(infile))
scan_eoln(infile);
if (fscanf(infile, "%lf", &x[i][j]) != 1) {
printf("The infile is of the wrong type\n");
exxit(-1);
}
if (replicates) {
if (eoln(infile))
scan_eoln(infile);
if (fscanf(infile, "%ld", &reps[i][j]) != 1) {
printf("The infile is of the wrong type\n");
exxit(-1);
}
} else
reps[i][j] = 1;
}
if (!skipit && skipother) {
x[j][i] = x[i][j];
reps[j][i] = reps[i][j];
}
if ((i == j) && (fabs(x[i][j]) > 0.000000001)) {
printf("\nERROR: diagonal element of row %ld of distance matrix ", i+1);
printf("is not zero.\n");
printf(" Is it a distance matrix?\n\n");
exxit(-1);
}
if ((j < i) && (fabs(x[i][j]-x[j][i]) > 0.000000001)) {
printf("ERROR: distance matrix is not symmetric:\n");
printf(" (%ld,%ld) element and (%ld,%ld) element are unequal.\n",
i+1, j+1, j+1, i+1);
printf(" They are %10.6f and %10.6f, respectively.\n",
x[i][j], x[j][i]);
printf(" Is it a distance matrix?\n\n");
exxit(-1);
}
}
}
scan_eoln(infile);
if (!printdata)
return;
for (i = 0; i < spp; i++) {
for (j = 0; j < nmlngth; j++)
putc(nayme[i][j], outfile);
putc(' ', outfile);
for (j = 1; j <= spp; j++) {
fprintf(outfile, "%10.5f", x[i][j - 1]);
if (replicates)
fprintf(outfile, " (%3ld)", reps[i][j - 1]);
if (j % columns == 0 && j < spp) {
putc('\n', outfile);
for (k = 1; k <= nmlngth + 1; k++)
putc(' ', outfile);
}
}
putc('\n', outfile);
}
putc('\n', outfile);
} /* inputdata */
void inputdata2(pointptr2 treenode)
{
/* input the names and character state data for species */
/* used in Mix & Penny */
long i, j, l;
char k;
Char charstate;
/* possible states are '0', '1', 'P', 'B', and '?' */
if (printdata)
headings(chars, "Characters", "----------");
for (i = 0; i < (chars); i++)
extras[i] = 0;
for (i = 1; i <= spp; i++) {
initname(i-1);
if (printdata) {
for (j = 0; j < nmlngth; j++)
putc(nayme[i - 1][j], outfile);
}
fprintf(outfile, " ");
for (j = 0; j < (words); j++) {
treenode[i - 1]->fulstte1[j] = 0;
treenode[i - 1]->fulstte0[j] = 0;
treenode[i - 1]->empstte1[j] = 0;
treenode[i - 1]->empstte0[j] = 0;
}
for (j = 1; j <= (chars); j++) {
k = (j - 1) % bits + 1;
l = (j - 1) / bits + 1;
do {
if (eoln(infile))
scan_eoln(infile);
charstate = gettc(infile);
if (charstate == '\n' || charstate == '\t')
charstate = ' ';
} while (charstate == ' ');
if (charstate == 'b') charstate = 'B';
if (charstate == 'p') charstate = 'P';
if (charstate != '0' && charstate != '1' && charstate != '?' &&
charstate != 'P' && charstate != 'B') {
printf("\n\nERROR: Bad character state: %c ",charstate);
printf("at character %ld of species %ld\n\n", j, i);
exxit(-1);
}
if (printdata) {
newline(outfile, j, 55, nmlngth + 3);
putc(charstate, outfile);
if (j % 5 == 0)
putc(' ', outfile);
}
if (charstate == '1') {
treenode[i-1]->fulstte1[l-1] =
((long)treenode[i-1]->fulstte1[l-1]) | (1L << k);
treenode[i-1]->empstte1[l-1] =
treenode[i-1]->fulstte1[l-1];
}
if (charstate == '0') {
treenode[i-1]->fulstte0[l-1] =
((long)treenode[i-1]->fulstte0[l-1]) | (1L << k);
treenode[i-1]->empstte0[l-1] =
treenode[i-1]->fulstte0[l-1];
}
if (charstate == 'P' || charstate == 'B')
extras[j-1] += weight[j-1];
}
scan_eoln(infile);
if (printdata)
putc('\n', outfile);
}
fprintf(outfile, "\n\n");
} /* inputdata2 */
void seqboot_inputdata()
{
/* input the names and sequences for each species */
long i, j, k, l, m, n, basesread, basesnew=0;
double x;
Char charstate;
boolean allread, done;
nodep = matrix_char_new(spp, sites);
j = nmlngth + (sites + (sites - 1) / 10) / 2 - 5;
if (j < nmlngth - 1)
j = nmlngth - 1;
if (j > 37)
j = 37;
interleaved = (interleaved && ((data == seqs) || (data == restsites)));
basesread = 0;
allread = false;
while (!allread) {
/* eat white space -- if the separator line has spaces on it*/
do {
charstate = gettc(infile);
} while (charstate == ' ' || charstate == '\t');
ungetc(charstate, infile);
if (eoln(infile))
scan_eoln(infile);
i = 1;
while (i <= spp) {
if ((interleaved && basesread == 0) || !interleaved)
initname(i-1);
j = interleaved ? basesread : 0;
done = false;
while (!done && !eoff(infile)) {
if (interleaved)
done = true;
while (j < sites && !(eoln(infile) ||eoff(infile))) {
charstate = gettc(infile);
if (charstate == '\n' || charstate == '\t')
charstate = ' ';
if (charstate == ' ' ||
(data == seqs && charstate >= '0' && charstate <= '9'))
continue;
uppercase(&charstate);
j++;
if (charstate == '.')
charstate = nodep[0][j-1];
nodep[i-1][j-1] = charstate;
}
if (interleaved)
continue;
if (j < sites)
scan_eoln(infile);
else if (j == sites)
done = true;
}
if (interleaved && i == 1)
basesnew = j;
scan_eoln(infile);
if ((interleaved && j != basesnew) || ((!interleaved) && j != sites)) {
printf("\n\nERROR: sequences out of alignment at site %ld", j+1);
printf(" of species %ld\n\n", i);
exxit(-1);
}
i++;
}
if (interleaved) {
basesread = basesnew;
allread = (basesread == sites);
} else
allread = (i > spp);
}
if (!printdata)
return;
} /* seqboot_inputdata */
void inputoptions()
{
/* input the information on the options */
long i;
if(justwts) {
if(firstset) {
scan_eoln(infile);
if (ancvar) {
inputancestors(anczero0, ancone0);
}
if (mixture) {
inputmixture(wagner0);
}
}
for (i = 0; i < (chars); i++)
weight[i] = 1;
inputweights(chars, weight, &weights);
for (i = 0; i < (words); i++) {
if (mixture)
wagner[i] = wagner0[i];
else if (allsokal)
wagner[i] = 0;
else
wagner[i] = (1L << (bits + 1)) - (1L << 1);
}
}
else {
if (!firstset) {
samenumsp(&chars, ith);
}
scan_eoln(infile);
for (i = 0; i < (chars); i++)
weight[i] = 1;
if (ancvar) {
inputancestors(anczero0, ancone0);
}
if (mixture) {
inputmixture(wagner0);
}
if (weights)
inputweights(chars, weight, &weights);
for (i = 0; i < (words); i++) {
if (mixture)
wagner[i] = wagner0[i];
else if (allsokal)
wagner[i] = 0;
else
wagner[i] = (1L << (bits + 1)) - (1L << 1);
}
}
for (i = 0; i < (chars); i++) {
if (!ancvar) {
anczero[i] = true;
ancone[i] = (((1L << (i % bits + 1)) & wagner[i / bits]) != 0);
} else {
anczero[i] = anczero0[i];
ancone[i] = ancone0[i];
}
}
noroot = true;
questions = false;
for (i = 0; i < (chars); i++) {
if (weight[i] > 0) {
noroot = (noroot && ancone[i] && anczero[i] &&
((((1L << (i % bits + 1)) & wagner[i / bits]) != 0)
|| threshold <= 2.0));
}
questions = (questions || (ancone[i] && anczero[i]));
threshwt[i] = threshold * weight[i];
}
} /* inputoptions */
void mexFunction(int nlhs, mxArray *plhs[], int nrhs, const mxArray *prhs[])
{ /* main program */
double *outputval, *inputval;
int mrows, ncols;
if (nrhs != 1) {
mexPrintf("Format for calling 'NEIGHBOR' executable:\n");
mexPrintf("\tNEIGHBOR( <outgrno> )\n");
mexErrMsgTxt("Wrong number of input args.");
}
/* Read in the argument: either a double or NULL '[]' */
if (((mxIsDouble(prhs[0]) != 1) & (!mxIsEmpty(prhs[0]))) == 1) {
mexErrMsgTxt("The argument should be a double.\n");
}
else if (((mxIsDouble(prhs[0]) == 1) & (!mxIsEmpty(prhs[0])) == 1)) {
mrows = mxGetM(prhs[0]);
ncols = mxGetN(prhs[0]);
if ((mrows * ncols) > 1)
mexErrMsgTxt("The argument should be a single value.\n");
else {
inputval = mxGetPr(prhs[0]);
outgrno = (long)*inputval;
}
}
else if (mxIsEmpty(prhs[0])) {
mexErrMsgTxt("numDouble = NULL\n");
}
else
mexErrMsgTxt("ERROR in <input_double> processing code.\n");
openfile(&infile,INFILE,"input file", "r","Neighbor",infilename);
openfile(&outfile,OUTFILE,"output file", "w","Neighbor",outfilename);
ibmpc = IBMCRT;
ansi = ANSICRT;
mulsets = false;
datasets = 1;
doinit();
if (trout){
openfile(&outtree,OUTTREE,"output tree file", "w","Neighbor",outtreename);
/* The next line is added to make a 'real' rooted tree -- James J. Cai 6/6/2006 */
putc('(', outtree);
}
ith = 1;
while (ith <= datasets) {
if (datasets > 1) {
fprintf(outfile, "Data set # %ld:\n",ith);
if (progress)
mexPrintf("Data set # %ld:\n",ith);
}
inputoptions();
maketree();
if (eoln(infile) && (ith < datasets))
scan_eoln(infile);
ith++;
}
FClose(infile);
FClose(outfile);
FClose(outtree);
freerest();
/* mexPrintf("Done.\n\n"); */
plhs[0] = mxCreateDoubleMatrix(1, 1, mxREAL);
outputval = mxGetPr(plhs[0]);
/* memcpy(output, 1.0, sizeof(double)); */
outputval[0] = 0;
return;
}
void maketree()
{
/* contruct the tree */
long nextsp,numtrees;
boolean succeeded=false;
long i, j, which;
if (usertree) {
inputdata(replicates, printdata, lower, upper, x, reps);
setuptree(&curtree, nonodes2);
for (which = 1; which <= spp; which++)
setuptipf(which, &curtree);
if (eoln(infile))
scan_eoln(infile);
openfile(&intree,INTREE,"input tree file","r",progname,intreename);
numtrees = countsemic(&intree);
if (numtrees > MAXNUMTREES) {
printf("\nERROR: number of input trees is read incorrectly from %s\n",
intreename);
exxit(-1);
}
if (treeprint) {
fprintf(outfile, "User-defined tree");
if (numtrees > 1)
putc('s', outfile);
fprintf(outfile, ":\n\n");
}
first = true;
which = 1;
while (which <= numtrees) {
treeread2 (intree, &curtree.start, curtree.nodep,
lengths, &trweight, &goteof, &haslengths, &spp,false,nonodes2);
nums = spp;
curtree.start = curtree.nodep[outgrno - 1]->back;
treevaluate();
printree(curtree.start, treeprint, false, false);
summarize(numtrees);
clear_connections(&curtree,nonodes2);
which++;
}
FClose(intree);
} else {
if (jumb == 1) {
inputdata(replicates, printdata, lower, upper, x, reps);
setuptree(&curtree, nonodes2);
setuptree(&priortree, nonodes2);
setuptree(&bestree, nonodes2);
if (njumble > 1) setuptree(&bestree2, nonodes2);
}
for (i = 1; i <= spp; i++)
enterorder[i - 1] = i;
if (jumble)
randumize(seed, enterorder);
nextsp = 3;
buildsimpletree(&curtree, nextsp);
curtree.start = curtree.nodep[enterorder[0] - 1]->back;
if (jumb == 1) numtrees = 1;
nextsp = 4;
if (progress) {
printf("Adding species:\n");
writename(0, 3, enterorder);
#ifdef WIN32
phyFillScreenColor();
#endif
}
while (nextsp <= spp) {
nums = nextsp;
buildnewtip(enterorder[nextsp - 1], &curtree, nextsp);
copy_(&curtree, &priortree);
bestree.likelihood = -99999.0;
curtree.start = curtree.nodep[enterorder[0] - 1]->back;
addtraverse(curtree.nodep[enterorder[nextsp - 1] - 1]->back,
curtree.start, true, &numtrees,&succeeded);
copy_(&bestree, &curtree);
if (progress) {
writename(nextsp - 1, 1, enterorder);
#ifdef WIN32
phyFillScreenColor();
#endif
}
if (global && nextsp == spp) {
if (progress) {
printf("Doing global rearrangements\n");
printf(" !");
for (j = spp; j < nonodes2; j++)
if ( (j - spp) % (( nonodes2 / 72 ) + 1 ) == 0 )
putchar('-');
printf("!\n");
printf(" ");
}
}
succeeded = true;
while (succeeded) {
succeeded = false;
curtree.start = curtree.nodep[enterorder[0] - 1]->back;
if (nextsp == spp && global)
globrearrange (&numtrees,&succeeded);
else{
rearrange(curtree.start,&numtrees,&nextsp,&succeeded);
}
if (global && ((nextsp) == spp) && progress)
printf("\n ");
}
if (global && nextsp == spp) {
putc('\n', outfile);
if (progress)
putchar('\n');
}
if (njumble > 1) {
if (jumb == 1 && nextsp == spp)
copy_(&bestree, &bestree2);
else if (nextsp == spp) {
if (bestree2.likelihood < bestree.likelihood)
copy_(&bestree, &bestree2);
}
}
if (nextsp == spp && jumb == njumble) {
if (njumble > 1) copy_(&bestree2, &curtree);
curtree.start = curtree.nodep[outgrno - 1]->back;
printree(curtree.start, treeprint, true, false);
summarize(numtrees);
}
nextsp++;
}
}
if (jumb == njumble && progress) {
printf("\nOutput written to file \"%s\"\n\n", outfilename);
if (trout) {
printf("Tree also written onto file \"%s\"\n", outtreename);
putchar('\n');
}
}
} /* maketree */
void seqboot_inputdata()
{
/* input the names and sequences for each species */
long i, j, k, l, m, n, basesread, basesnew=0;
double x;
Char charstate;
boolean allread, done;
if (data == genefreqs) {
nodef = matrix_double_new(spp, sites);
} else {
nodep = matrix_char_new(spp, sites);
}
j = nmlngth + (sites + (sites - 1) / 10) / 2 - 5;
if (j < nmlngth - 1)
j = nmlngth - 1;
if (j > 37)
j = 37;
if (printdata) {
fprintf(outfile, "\nBootstrapping algorithm, version %s\n\n\n",VERSION);
if (bootstrap) {
if (blocksize > 1) {
if (regular)
fprintf(outfile, "Block-bootstrap with block size %ld\n\n", blocksize);
else
fprintf(outfile, "Partial (%2.0f%%) block-bootstrap with block size %ld\n\n",
100*fracsample, blocksize);
} else {
if (regular)
fprintf(outfile, "Bootstrap\n\n");
else
fprintf(outfile, "Partial (%2.0f%%) bootstrap\n\n", 100*fracsample);
}
} else {
if (jackknife) {
if (regular)
fprintf(outfile, "Delete-half Jackknife\n\n");
else
fprintf(outfile, "Delete-%2.0f%% Jackknife\n\n", 100*(1.0-fracsample));
} else {
if (permute) {
fprintf(outfile, "Species order permuted separately for each");
if (data == genefreqs)
fprintf(outfile, " locus\n\n");
if (data == seqs)
fprintf(outfile, " site\n\n");
if (data == morphology)
fprintf(outfile, " character\n\n");
if (data == restsites)
fprintf(outfile, " site\n\n");
}
else {
if (ild) {
if (data == genefreqs)
fprintf(outfile, "Locus");
if (data == seqs)
fprintf(outfile, "Site");
if (data == morphology)
fprintf(outfile, "Character");
if (data == restsites)
fprintf(outfile, "Site");
fprintf(outfile, " order permuted\n\n");
} else {
if (lockhart)
if (data == genefreqs)
fprintf(outfile, "Locus");
if (data == seqs)
fprintf(outfile, "Site");
if (data == morphology)
fprintf(outfile, "Character");
if (data == restsites)
fprintf(outfile, "Site");
fprintf(outfile, " order permuted separately for each species\n\n");
}
}
}
}
if (data == genefreqs)
fprintf(outfile, "%3ld species, %3ld loci\n\n", spp, loci);
else {
fprintf(outfile, "%3ld species, ", spp);
if (data == seqs)
fprintf(outfile, "%3ld sites\n\n", sites);
else if (data == morphology)
fprintf(outfile, "%3ld characters\n\n", sites);
else if (data == restsites)
fprintf(outfile, "%3ld sites\n\n", sites);
}
fprintf(outfile, "Name");
for (i = 1; i <= j; i++)
putc(' ', outfile);
fprintf(outfile, "Data\n");
fprintf(outfile, "----");
for (i = 1; i <= j; i++)
putc(' ', outfile);
fprintf(outfile, "----\n\n");
}
interleaved = (interleaved && ((data == seqs) || (data == restsites)));
if (data == genefreqs) {
for (i = 1; i <= (spp); i++) {
initname(i - 1);
j = 1;
while (j <= sites && !eoff(infile)) {
if (eoln(infile))
scan_eoln(infile);
if ( fscanf(infile, "%lf", &x) != 1) {
printf("ERROR: Invalid value for locus %ld of species %ld\n", j, i);
exxit(-1);
} else if ((unsigned)x > 1.0) {
printf("GENE FREQ OUTSIDE [0,1] in species %ld\n", i);
exxit(-1);
} else {
nodef[i - 1][j - 1] = x;
j++;
}
}
scan_eoln(infile);
}
return;
}
basesread = 0;
allread = false;
while (!allread) {
/* eat white space -- if the separator line has spaces on it*/
do {
charstate = gettc(infile);
} while (charstate == ' ' || charstate == '\t');
ungetc(charstate, infile);
if (eoln(infile))
scan_eoln(infile);
i = 1;
while (i <= spp) {
if ((interleaved && basesread == 0) || !interleaved)
initname(i-1);
j = interleaved ? basesread : 0;
done = false;
while (!done && !eoff(infile)) {
if (interleaved)
done = true;
while (j < sites && !(eoln(infile) ||eoff(infile))) {
charstate = gettc(infile);
if (charstate == '\n' || charstate == '\t')
charstate = ' ';
if (charstate == ' ' ||
(data == seqs && charstate >= '0' && charstate <= '9'))
continue;
uppercase(&charstate);
j++;
if (charstate == '.')
charstate = nodep[0][j-1];
nodep[i-1][j-1] = charstate;
}
if (interleaved)
continue;
if (j < sites)
scan_eoln(infile);
else if (j == sites)
done = true;
}
if (interleaved && i == 1)
basesnew = j;
scan_eoln(infile);
if ((interleaved && j != basesnew) || ((!interleaved) && j != sites)){
printf("\n\nERROR: sequences out of alignment at site %ld", j+1);
printf(" of species %ld\n\n", i);
exxit(-1);}
i++;
}
if (interleaved) {
basesread = basesnew;
allread = (basesread == sites);
} else
allread = (i > spp);
}
if (!printdata)
return;
if (data == genefreqs)
m = (sites - 1) / 8 + 1;
else
m = (sites - 1) / 60 + 1;
for (i = 1; i <= m; i++) {
for (j = 0; j < spp; j++) {
for (k = 0; k < nmlngth; k++)
putc(nayme[j][k], outfile);
fprintf(outfile, " ");
if (data == genefreqs)
l = i * 8;
else
l = i * 60;
if (l > sites)
l = sites;
if (data == genefreqs)
n = (i - 1) * 8;
else
n = (i - 1) * 60;
for (k = n; k < l; k++) {
if (data == genefreqs)
fprintf(outfile, "%8.5f", nodef[j][k]);
else {
if (j + 1 > 1 && nodep[j][k] == nodep[0][k])
charstate = '.';
else
charstate = nodep[j][k];
putc(charstate, outfile);
if ((k + 1) % 10 == 0 && (k + 1) % 60 != 0)
putc(' ', outfile);
}
}
putc('\n', outfile);
}
putc('\n', outfile);
}
putc('\n', outfile);
} /* seqboot_inputdata */
void restdist_inputdata()
{
/* read the species and sites data */
long i, j, k, l, sitesread = 0, sitesnew = 0;
Char ch;
boolean allread, done;
if (printdata)
putc('\n', outfile);
j = nmlngth + (sites + (sites - 1) / 10) / 2 - 5;
if (j < nmlngth - 1)
j = nmlngth - 1;
if (j > 39)
j = 39;
if (printdata) {
fprintf(outfile, "Name");
for (i = 1; i <= j; i++)
putc(' ', outfile);
fprintf(outfile, "Sites\n");
fprintf(outfile, "----");
for (i = 1; i <= j; i++)
putc(' ', outfile);
fprintf(outfile, "-----\n\n");
}
sitesread = 0;
allread = false;
while (!(allread)) {
/* eat white space -- if the separator line has spaces on it*/
do {
ch = gettc(infile);
} while (ch == ' ' || ch == '\t');
ungetc(ch, infile);
if (eoln(infile))
scan_eoln(infile);
i = 1;
while (i <= spp ) {
if ((interleaved && sitesread == 0) || !interleaved)
initname(i - 1);
if (interleaved)
j = sitesread;
else
j = 0;
done = false;
while (!done && !eoff(infile)) {
if (interleaved)
done = true;
while (j < sites && !(eoln(infile) || eoff(infile))) {
ch = gettc(infile);
if (ch == '\n' || ch == '\t')
ch = ' ';
if (ch == ' ')
continue;
uppercase(&ch);
if (ch != '1' && ch != '0' && ch != '+' && ch != '-' && ch != '?') {
printf(" ERROR -- Bad symbol %c",ch);
printf(" at position %ld of species %ld\n", j+1, i);
exxit(-1);
}
if (ch == '1')
ch = '+';
if (ch == '0')
ch = '-';
j++;
y[i - 1][j - 1] = ch;
}
if (interleaved)
continue;
if (j < sites)
scan_eoln(infile);
else if (j == sites)
done = true;
}
if (interleaved && i == 1)
sitesnew = j;
scan_eoln(infile);
if ((interleaved && j != sitesnew ) || ((!interleaved) && j != sites)){
printf("ERROR: SEQUENCES OUT OF ALIGNMENT\n");
exxit(-1);}
i++;
}
if (interleaved) {
sitesread = sitesnew;
allread = (sitesread == sites);
} else
allread = (i > spp);
}
if (printdata) {
for (i = 1; i <= ((sites - 1) / 60 + 1); i++) {
for (j = 0; j < spp; j++) {
for (k = 0; k < nmlngth; k++)
putc(nayme[j][k], outfile);
fprintf(outfile, " ");
l = i * 60;
if (l > sites)
l = sites;
for (k = (i - 1) * 60 + 1; k <= l; k++) {
putc(y[j][k - 1], outfile);
if (k % 10 == 0 && k % 60 != 0)
putc(' ', outfile);
}
putc('\n', outfile);
}
putc('\n', outfile);
}
putc('\n', outfile);
}
} /* restdist_inputdata */