1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139 140 141 142 143 144 145 146 147 148 149 150 151 152 153 154 155 156 157 158 159 160 161 162 163 164 165 166 167 168 169 170 171 172 173 174 175 176 177 178 179 180 181 182 183 184 185 186 187 188 189 190 191 192 193 194 195 196 197 198 199 200 201 202 203 204 205 206 207 208 209 210 211 212 213 214 215 216 217 218 219 220 221 222 223 224 225 226 227 228 229 230 231 232 233 234 235 236 237 238 239 240 241 242 243 244 245 246 247 248 249 250 251 252 253 254 255 256 257 258 259 260 261 262 263 264 265 266 267 268 269 270 271 272 273 274 275 276 277 278 279 280 281 282 283 284 285 286 287 288 289 290 291 292 293 294 295 296 297 298 299 300 301 302 303 304 305 306 307 308 309 310 311 312 313 314 315 316 317 318 319 320 321 322 323 324 325 326 327 328 329 330 331 332 333 334 335 336 337 338 339 340 341 342 343 344 345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371 372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422 423 424 425 426 427 428 429 430 431 432 433 434 435 436 437 438 439 440 441 442 443 444 445 446 447 448 449 450 451 452 453 454 455 456 457 458 459 460 461 462 463 464 465 466 467 468 469 470 471 472 473 474 475 476 477 478 479 480 481 482 483 484 485 486 487 488 489 490 491 492 493 494 495 496 497 498 499 500 501 502 503 504 505 506 507 508 509 510 511 512 513 514 515 516 517 518 519 520 521 522 523 524 525 526 527 528 529 530 531 532 533 534 535 536 537 538 539 540 541 542 543 544 545 546 547 548 549 550 551 552 553 554 555 556 557 558 559 560 561 562 563 564 565 566 567 568 569 570 571 572 573 574 575 576 577 578 579 580 581 582 583 584 585 586 587 588 589 590 591 592 593 594 595 596 597 598 599 600 601 602 603 604 605 606 607 608 609 610 611 612 613 614 615 616 617 618 619 620 621 622 623 624 625 626 627 628 629 630 631 632 633 634 635 636 637 638 639 640 641 642 643 644 645 646 647 648 649 650 651 652 653 654 655 656 657 658 659 660 661 662 663 664 665 666 667 668 669 670 671 672 673 674 675 676 677 678 679 680 681 682 683 684 685 686 687 688 689 690 691 692 693 694 695 696 697 698 699 700 701 702 703 704 705 706 707 708 709 710 711 712 713 714 715 716 717 718 719 720 721 722 723 724 725 726 727 728 729 730 731 732 733 734 735 736 737 738 739 740 741 742 743 744 745 746 747 748 749 750 751 752 753 754 755 756 757 758 759 760 761 762 763 764 765 766 767 768 769 770 771 772 773 774 775 776 777 778 779 780 781 782 783 784 785 786 787 788 789 790 791 792 793 794 795 796 797 798 799 800 801 802 803 804 805 806 807 808 809 810 811 812 813 814 815 816 817 818 819 820 821 822 823 824 825 826 827 828 829 830 831 832 833 834 835 836 837 838 839 840 841 842 843 844 845 846 847 848 849 850 851 852 853 854 855 856 857 858 859 860 861 862 863 864 865 866 867 868 869 870 871 872 873 874 875 876 877 878 879 880 881 882 883 884 885 886 887 888 889 890 891 892 893 894 895 896 897 898 899 900 901 902 903 904 905 906 907 908 909 910 911 912 913 914 915 916 917 918 919 920 921 922 923 924 925 926 927 928 929 930 931 932 933 934 935 936 937 938 939 940 941 942 943 944 945 946 947 948 949 950 951 952 953 954 955 956 957 958 959 960 961 962 963 964 965 966 967 968 969 970 971 972 973 974 975 976 977 978 979 980 981 982 983 984 985 986 987 988 989 990 991 992 993 994 995 996 997 998 999 1000 1001 1002 1003 1004 1005 1006 1007 1008 1009 1010 1011 1012 1013 1014 1015 1016 1017 1018 1019 1020 1021 1022 1023 1024 1025 1026 1027 1028 1029 1030 1031 1032 1033 1034 1035 1036 1037 1038 1039 1040 1041 1042 1043 1044 1045 1046 1047 1048 1049 1050 1051 1052 1053 1054 1055 1056 1057 1058 1059 1060 1061 1062 1063 1064 1065 1066 1067 1068 1069 1070 1071 1072 1073 1074 1075 1076 1077 1078 1079 1080 1081 1082 1083 1084 1085 1086 1087 1088 1089 1090 1091 1092 1093 1094 1095 1096 1097 1098 1099 1100 1101 1102 1103 1104 1105 1106 1107 1108 1109 1110 1111 1112 1113 1114 1115 1116 1117 1118 1119 1120 1121 1122 1123 1124 1125 1126 1127 1128 1129 1130 1131 1132 1133 1134 1135 1136 1137 1138 1139 1140 1141 1142 1143 1144 1145 1146 1147 1148 1149 1150 1151 1152 1153 1154 1155 1156 1157 1158 1159 1160 1161 1162 1163 1164 1165 1166 1167 1168 1169 1170 1171 1172 1173 1174 1175 1176 1177 1178 1179 1180 1181 1182 1183 1184 1185 1186 1187 1188 1189 1190 1191 1192 1193 1194 1195 1196 1197 1198 1199 1200 1201 1202 1203 1204 1205 1206 1207 1208 1209 1210 1211 1212 1213 1214 1215 1216 1217 1218 1219 1220 1221 1222 1223 1224 1225 1226 1227 1228 1229 1230 1231 1232 1233 1234 1235 1236 1237 1238 1239 1240 1241 1242 1243 1244 1245 1246 1247 1248 1249 1250 1251 1252 1253 1254 1255 1256 1257 1258 1259 1260 1261 1262 1263 1264 1265 1266 1267 1268 1269 1270 1271 1272 1273 1274 1275 1276 1277 1278 1279 1280 1281 1282 1283 1284 1285 1286 1287 1288 1289 1290 1291 1292 1293 1294 1295 1296 1297 1298 1299 1300 1301 1302 1303 1304 1305 1306 1307 1308 1309 1310 1311 1312 1313 1314 1315 1316 1317 1318 1319 1320 1321 1322 1323 1324 1325 1326 1327 1328 1329 1330 1331 1332 1333 1334 1335 1336 1337 1338 1339 1340 1341 1342 1343 1344 1345 1346 1347 1348 1349 1350 1351 1352 1353 1354 1355 1356 1357 1358 1359 1360 1361 1362 1363 1364 1365 1366 1367 1368 1369 1370 1371 1372 1373 1374 1375 1376 1377 1378 1379 1380 1381 1382 1383 1384 1385 1386 1387 1388 1389 1390 1391 1392 1393 1394 1395 1396 1397 1398 1399 1400 1401 1402 1403 1404 1405 1406 1407 1408 1409 1410 1411 1412 1413 1414 1415 1416 1417 1418 1419 1420 1421 1422 1423 1424 1425 1426 1427 1428 1429 1430 1431 1432 1433 1434 1435 1436 1437 1438 1439 1440 1441 1442 1443 1444 1445 1446 1447 1448 1449 1450 1451 1452 1453 1454 1455 1456 1457 1458 1459 1460 1461 1462 1463 1464 1465 1466 1467 1468 1469 1470 1471 1472 1473 1474 1475 1476 1477 1478 1479 1480 1481 1482 1483 1484 1485 1486 1487 1488 1489 1490 1491 1492 1493 1494 1495 1496 1497 1498 1499 1500 1501 1502 1503 1504 1505 1506 1507 1508 1509 1510 1511 1512 1513 1514 1515 1516 1517 1518 1519 1520 1521 1522 1523 1524 1525 1526 1527 1528 1529 1530 1531 1532 1533 1534 1535 1536 1537 1538 1539 1540 1541 1542 1543 1544 1545 1546 1547 1548 1549 1550 1551 1552 1553 1554 1555 1556 1557 1558 1559 1560 1561 1562 1563 1564 1565 1566 1567 1568 1569 1570 1571 1572 1573 1574 1575 1576 1577 1578 1579 1580 1581 1582 1583 1584 1585 1586 1587 1588 1589 1590 1591 1592 1593 1594 1595 1596 1597 1598 1599 1600 1601 1602 1603 1604 1605 1606 1607 1608 1609 1610 1611 1612 1613 1614 1615 1616 1617 1618 1619 1620 1621 1622 1623 1624 1625 1626 1627 1628 1629 1630 1631 1632 1633 1634 1635 1636 1637 1638 1639 1640 1641 1642 1643 1644 1645 1646 1647 1648 1649 1650
|
/*======================================================================*/
/* (C) Copyright 1991-2000 by Fred Hutchinson Cancer Research Center */
/* motomat.c Block extension and assembly program */
/* USE: motomat <input file> <cutoff score> <cluster %> <drop sd*10|asis> */
/* EG: motomat PS00094.mot 1 80 -10 */
/* input file = .mot file from motifj.c (this is a binary file) */
/* cutoff score = average PAM score for all possible pairs of */
/* amino acids in a column of the block (0-2500). */
/* If the cutoff score = 1, motomat will determine the best */
/* block extension. This is the recommended choice. If the */
/* cutoff score =2, lots of intermediate output is produced. */
/* cluster % = Percent identical AAs to cluster seqs in .blk file*/
/* drop sd = sd of blocks score * 10, blocks scoring lower will */
/* be dropped and not assembled. EG -25 => -2.5 sds */
/* If drop sd == asis, blocks will not be merged, dropped or */
/* extended; duplicate blocks will be dropped */
/* outputs blocks with the extension ".blk" and file name made */
/* up of the 1st 7 chars of the input file title plus one */
/* character for each assembled block; EG PS00094A.blk */
/* Reference: S. Henikoff & J. Henikoff, NAR 19:23, 6565-6572 (1991)
See PROTOMAT documentation file for description of algorithms. */
/*----------------------------------------------------------------------
6/20/90 J. Henikoff
>>>>>>>>>>>>>>>>>>>>> Blocks 9.0 <<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<
6/29/97 Changed execlp() to use argv[0]
10/21/97 Round proportion of sequences instead of truncate in best_path()
10/28/97 Added pb_weights() to score_block() & score_cols()
11/ 8/97 Make min drop score == highpass value
11/10/97 Weight blocks in best_path() by nident instead of nmotif
Fixed bug in check_pos()
Added list of seqs not in best path to print_best()
11/11/97 Weight blocks in best_path() by nmotif from motifj
12/30/97 Added reduce_overlap()
1/19/98 Weight blocks in best_path() by sqrt(nmotif) instead of nmotif.
1/21/98 Fixed argv[0] error in restart(). Removed pb_weights() error message.
1/25/98 Moved reduce_overlap() to after blocks are dropped for low score.
1/28/98 Fixed reduce_overlap() bug, wrong block trimmed
New option (-asis) => don't drop any blocks, don't extend,
don't reduce overlap, etc.
2/12/98.1 Added version date (VerDate[]). Drop blocks < 4 wide.
Don't shorten block in reduce_overlap() if resulting block
will be < MIN_DOMAIN_WIDTH=10 wide.
Fixed overcounting of block->nident in score_block()
Added Overlap = #times reduce_overlap() shortened a block to
motomat.err
2/16/98.1 Drop blocks < 5 wide.
2/28/98.1 Modified trim_block() to never make block narrower.
6/20/99.1 Longer sequence names (SNAMELEN=18)
7/ 6/99.1 Longer AC names (MAXAC=10)
3/ 3/00.1 Memory bug fix
6/ 6/00.1 Fix blk_filename for BlockMaker; save all blocks in one file
save_block();
12/23/06.1 Longer sequence names (SNAMELEN=20)
==========================================================================*/
#include <math.h>
#include <unistd.h>
#include <ctype.h>
#include "motifj.h"
/*--------------------Function prototypes--------------------------------*/
void dump_data();
struct merged_motif *init_mm();
void merge_blocks();
int check_align();
int check_pos();
int check_overlap();
void dup_blocks();
void reduce_overlap();
void trim_block();
void extend_block();
void print_block();
void print_head();
void save_block();
void cluster_seqs();
void print_seq();
void write_seq();
void score_cols();
void score_block();
void pb_weights();
void restart();
/* Shotgun assembly stuff; code is in motomat2.c */
int prune_blocks();
void order_blocks();
int tempcmp();
void build_dag();
void find_paths();
void next_arc();
struct path *makepath();
struct block_list *makebllist();
void ins_path();
struct path *copypath();
void free_path();
struct matrix *makematrix();
void follow_arcs();
void ins_bllist();
struct path *best_paths();
void best_path();
void check_seqs();
void print_best();
void print_path();
void save_plot();
int left();
int right();
/*---------------Routines from motmisc.c -----------------------------*/
char *num_to_aachar();
int aachar_to_num();
void getscore();
void kr_itoa();
struct split_name *split_names();
/*------------------Global vars------------------------------------------*/
char VerDate[12] = " 6/ 6/00.1";
int Debug = NO; /* Flag for debugging print-out */
int AsIs = NO; /* if argv[5] exists */
int MinScore; /* Minimum column score for block extension */
int ClThres; /* Clustering threshold (%) */
int DropScore; /* # of sds from mean before block is dropped */
long TopScore; /* Actual drop score */
char Mot_Filename[FNAMELEN]; /* Name of the .mot input file */
char Blk_Filename[FNAMELEN]; /* Name of the .blk output file */
int Version, RunType, Signif, Dups, Distance; /* motifj parameters */
int RSignif; /* reduces significance */
int NumSeqs; /* number of sequences */
int Total_Motifs; /* # motifs from motifj */
int Overlap; /* # reduce_overlap() */
char AC[MAXAC+1], ACName[MAXLINE], ID[MAXLINE], DE[MAXLINE];
int MinDist, MaxDist;
int Gibbs = NO; /* Gibbs input flag */
struct matrix *Dag[RELEVANT_MOTIFS][RELEVANT_MOTIFS]; /* Graph */
struct score *SMatrix; /* Scoring matrix */
char Argv0[MAXLINE]; /* Save argv[0] for restart */
FILE *Fblk; /* .blks output file */
/*=======================================================================*/
int main(argc, argv)
int argc;
char *argv[];
{
FILE *mot, *err;
char title[MAXLINE], ctemp[MAXLINE], *ptr;
int done, i, j, b, c, totb, motauto, maxdups;
int lastscore, lastthres, lastdrop;
int pseqs[MAXSEQS];
double ftemp, fmean, fsdev;
struct sequences *seqs;
struct motif_struct *motif;
struct merged_motif *blocks;
struct path *paths;
struct split_name *motsplit;
printf("\nMOTOMAT %s: ", VerDate);
printf("(C) Copyright 1991 by Fred Hutchinson Cancer Research Center");
printf("\nPlease cite: S.Henikoff & J.Henikoff, NAR 19:6565-6572 (1991).");
/*-------Get the scoring matrix and high pass filter values --------*/
SMatrix = (struct score *) malloc(sizeof(struct score));
if (SMatrix != NULL) getscore(SMatrix);
else {printf("\nCANNOT ALLOCATE SMatrix\n"); exit(-1);}
strcpy(Argv0, argv[0]);
/*----------------Get the input file -------------------------------*/
if (argc > 1)
strcpy(Mot_Filename, argv[1]);
else
{
printf("\n");
system("ls -al *.mot");
printf("\nEnter name of file containing motifs: ");
gets(Mot_Filename);
}
ptr = strrchr(Mot_Filename, '.'); /* check for extension */
if (ptr == NULL) strcat(Mot_Filename, ".mot");
if ( (mot=fopen(Mot_Filename, "rb")) == NULL)
{
printf("\nCannot open file %s\n", Mot_Filename);
exit(-1);
}
motsplit=split_names(Mot_Filename); /* split up name */
Blk_Filename[0] = '\0';
strncat(Blk_Filename, Mot_Filename+motsplit->dir_len, motsplit->name_len);
Blk_Filename[motsplit->name_len] = '\0';
/*----------------Read the data file ------------------------------------*/
title[0] = '\0';
NumSeqs = Total_Motifs = -1;
fread(&Version, sizeof(int), 1, mot);
if (Version != VERSION)
{
printf("\n\nInput file was created using wrong motifj version!\n");
fclose(mot);
exit(-1);
}
fread(&RunType, sizeof(int), 1, mot);
fread(&Signif, sizeof(int), 1, mot);
fread(&Dups, sizeof(int), 1, mot);
fread(&Distance, sizeof(int), 1, mot);
fread(&NumSeqs, sizeof(int), 1, mot); /* # sequences */
fread(&Total_Motifs, sizeof(int), 1, mot); /* # motifs */
seqs = (struct sequences *) malloc(sizeof (struct sequences));
seqs->num = NumSeqs;
seqs->len = (int *) malloc(NumSeqs * sizeof (int));
fread(seqs->len, sizeof(int), NumSeqs, mot); /* sequence lengths */
seqs->totlen = 0;
seqs->offlen = (int *) malloc(NumSeqs * sizeof (int));
for (i=0; i<NumSeqs; i++)
{
*(seqs->offlen+i) = seqs->totlen; /* offset to ith sequence */
seqs->totlen += *(seqs->len+i);
}
/* sequence names */
seqs->name = (char *) malloc(SNAMELEN * NumSeqs * sizeof (char));
fread(seqs->name, SNAMELEN*sizeof(char), NumSeqs, mot);
seqs->seq = (char *) malloc(seqs->totlen * sizeof(char));
fread(seqs->seq, sizeof(char), seqs->totlen, mot); /* sequences */
if (Total_Motifs > 0)
{
motif = (struct motif_struct *)
malloc(Total_Motifs * sizeof(struct motif_struct));
/*--- For Gibbs, Total_Motifs is the maximum number, may be fewer */
for (i=0; i<Total_Motifs; i++) motif[i].domain = -1;
if (RunType == 9) /* output from gibbs */
{
Gibbs = YES;
if (!feof(mot)) fread(title, 100*sizeof(char), 1, mot);
printf("\nAssuming input from Gibbs\n");
i = 0;
while (!feof(mot) && i<Total_Motifs)
{
fread(&motif[i].domain, sizeof(int), 1, mot);
fread(&motif[i].freq, sizeof(int), 1, mot);
for (j=0; j< motif[i].freq; j++)
{
fread(&motif[i].seq_no[j], sizeof(int), 1, mot);
motif[i].seq_no[j] -= 1;
fread(&motif[i].pos[j], sizeof(int), 1, mot);
motif[i].pos[j] -= 1;
}
motif[i].dups = motif[i].score = 0;
motif[i].mots = 1;
motif[i].distance1 = 0;
motif[i].aa1 = motif[i].aa2 = motif[i].aa3 = 'X';
motif[i].distance2 = motif[i].domain - 1;
i++;
}
if (i < Total_Motifs) Total_Motifs = i;
}
else
{
fread(motif, sizeof(struct motif_struct), Total_Motifs, mot);
if (!feof(mot)) fread(title, MAXLINE*sizeof(char), 1, mot);
}
}
else motif = NULL;
fclose(mot);
/*-------- Beware of incomplete records ---------------------*/
i = 0;
while (i < Total_Motifs && motif[i].domain > 0) i++;
if (i < Total_Motifs) Total_Motifs = i;
printf("\n%s\n", title);
printf("\n%d blocks read", Total_Motifs);
printf(" using parameters [%d, %d, %d].", Signif, Dups, Distance);
printf(" %d sequences.", NumSeqs);
/*--------------------------------------------------------------------*/
/* Get AC, ID, DE from title line */
/*-- General form is >ACxxxxx ;ID;Description;$ -------*/
ACName[0] = '\0';
strcpy(ID, "none"); strcpy(DE, "none");
/* title is missing or too short */
if (!(strlen(title)) || title[0] != '>')
{
strcat(ACName, Blk_Filename); /* use the filename as AC */
}
else
{
ptr = strtok(title+1, ";");
if (ptr != NULL)
{
strcpy(ACName, ptr);
ptr = strtok(NULL, ";");
if (ptr != NULL)
{
strcpy(ID, ptr);
ptr = strtok(NULL, ";");
if (ptr != NULL)
{
strcpy(DE, ptr);
}
}
}
else
{
strcpy(ACName, title+1);
}
}
ACName[MAXAC-1] = '\0';
for (i=0; i<MINAC; i++)
{
if (!isalnum(ACName[i])) ACName[i] = 'x';
}
i=MINAC; done = NO;
while (!done && i < MAXAC-1)
{
if (ACName[i] == ' ' || ACName[i] == ';')
{ ACName[i] = '\0'; done = YES; }
else { i++; }
}
/*------------------Get the parameters ----------------------------------*/
if (argc <= 3) RunType = 0; /* ask for the args */
else RunType = 3; /* don't ask for args */
motauto = NO; /* MinScore==1 is a flag for auto parameter selection */
if (argc > 2)
{
strcpy(ctemp, argv[2]);
MinScore = atoi(ctemp);
}
if (MinScore == 2)
{
Debug = YES; MinScore = 1;
}
if (MinScore == 1) /* Automatic determination of parameters */
{
motauto = YES; RunType = 3;
}
else if (MinScore < 100 || MinScore > 2500) MinScore = MINSCORE;
lastscore = MinScore;
if (argc > 3) /* Clustering threshold percentage */
{
strcpy(ctemp, argv[3]);
ClThres = atoi(ctemp);
}
if (ClThres <= 0 || ClThres > 100) ClThres = CLTHRES;
lastthres = ClThres;
if (argc > 4) /* Block drop score stand. dev. */
{
if (strcasecmp(argv[4], "asis") == 0) AsIs = YES;
else DropScore = atoi(argv[4]);
}
else if (motauto)
{
DropScore = -10; /* default = -1sd */
}
if (DropScore < -30 || DropScore > 30) DropScore = DROPSCORE;
lastdrop = DropScore;
printf("\nProcessing blocks in %s using a cut-off score of %d ",
Mot_Filename, MinScore);
printf("\n and a clustering threshold of %d percent.", ClThres);
/*----------------------Quit if nothing to do -------------------------*/
if (Debug)
dump_data(motif, seqs);
if (Total_Motifs < 1)
{
printf("\nQuitting becasuse no motifs to work on.\n");
exit(0);
}
/*-------------------Redirect standard error output --------------------*/
err = freopen("motomat.err", "a", stderr);
/*------------------Initialize sequences to save to all-----------------*/
for (i=0; i<NumSeqs; i++)
pseqs[i] = YES;
/*----------------------Create and merge the blocks --------------------*/
blocks = init_mm(motif);
free(motif); /* Done with motifs now */
/*---------------------Merge motifs if not from Gibbs ------------------*/
if (!Gibbs && !AsIs)
{
printf("\n\nMerging raw blocks...");
merge_blocks(blocks, 0);
}
/*-------Adjust the block width and recalculate the column scores -------*/
/*-------Do this over MAX_MERGE_WIDTH-----------------------------------*/
printf("\nRecalculating block column scores");
for (b=0; b<Total_Motifs; b++)
if (blocks[b].nmotif > 0)
{
if (!AsIs && blocks[b].domain+1 < MAX_MERGE_WIDTH)
{
blocks[b].loffset = blocks[b].t_loffset =
(int) (blocks[b].loffset +
(MAX_MERGE_WIDTH-blocks[b].domain-1)/2);
blocks[b].domain = blocks[b].t_domain = MAX_MERGE_WIDTH-1;
}
printf(".");
score_cols(&blocks[b], seqs);
}
/*--------- Extend, cluster & print out the top merged blocks -----------*/
printf("\nExtending blocks...");
totb=0; /* total surviving blocks */
RSignif=Signif; /* reduced significance */
sprintf(ctemp, "%s.blks", Blk_Filename);
if ((Fblk=fopen(ctemp, "w+t")) == NULL)
{
printf("\nCannot open %s\n", ctemp);
exit(-1);
}
printf("Saving blocks to %s\n", ctemp);
for (b=0; b<Total_Motifs; b++)
{
printf("\nBlock %d: %.1s%.1s%.1s",
b, num_to_aachar(blocks[b].aa[0]),
num_to_aachar(blocks[b].aa[1]),
num_to_aachar(blocks[b].aa[2]));
if (blocks[b].nmotif <= 0) { printf(" has been merged.");}
else
{
if (!AsIs)
{
do {
if (RunType == 0 || RunType == 2)
{
printf("\nEnter minimum block extension column score");
printf(" or 1 for best block [100-2500 or 1; %d]: ",
lastscore);
gets(ctemp);
MinScore = atoi(ctemp);
if ((MinScore > 1 && MinScore < 100) ||
MinScore <= 0 || MinScore > 2500)
MinScore = lastscore;
lastscore = MinScore;
printf("Enter clustering identity threshold [1-100; %d]: ",
lastthres);
gets(ctemp);
ClThres = atoi(ctemp);
if (ClThres <= 0 || ClThres > 100) ClThres = lastthres;
lastthres = ClThres;
printf("\nProcessing block %d in %s using a cut-off score of %d ",
b, Mot_Filename, MinScore);
printf("\n and a clustering threshold of %d percent.", ClThres);
} /* end of runtype = 0 or 2 */
totb++;
if (MinScore == 1)
{
extend_block(&blocks[b]);
if (blocks[b].t_domain+1 < MIN_DOMAIN_WIDTH)
{
MinScore = SMatrix->highpass;
while (blocks[b].t_domain+1 < MIN_DOMAIN_WIDTH &&
MinScore >= (SMatrix->highpass - 50))
{
printf("\nExtended width was %d",
blocks[b].t_domain+1);
printf(", re-extending with cutoff score of %d.",
MinScore);
trim_block(&blocks[b]);
MinScore -= 50;
}
MinScore = 1;
}
if (blocks[b].t_domain+1 < 5)
{
blocks[b].dropped = YES;
printf("\nBlock %d has been dropped for low width:", b);
print_head(&blocks[b]);
}
} /* end of if MinScore==1 */
else
{
trim_block(&blocks[b]);
}
if (RunType==0 || RunType==2)
{
score_block(&blocks[b], seqs);
cluster_seqs(&blocks[b], seqs);
print_block(&blocks[b], seqs);
printf("\n\nType S[ENTER] to save this block,");
printf(" or N[ENTER] to skip to next block,");
printf("\nor anything else[ENTER] to re-extend this block...");
c=getchar(); getchar(); /* 2nd get is for ENTER */
if (c == 's' || c == 'S')
{
save_block(&blocks[b], seqs, pseqs);
c = 'N';
}
}
} while ((RunType==0 || RunType==2) && c != 'N' && c != 'n');
} /* end of if !AsIs */
} /* end of else not merged */
} /* end of for b */
/*-----------Merge extended blocks if not from Gibbs -----------*/
if (!Gibbs && !AsIs)
{
printf("\n\nMerging extended blocks....");
merge_blocks(blocks, 1);
}
else
{
printf("\n\nRemoving duplicate blocks....");
dup_blocks(blocks);
}
/*---------------------------------------------------------------------*/
/* Shotgun assembly of all unmerged, high-scoring blocks. */
/*---------------------------------------------------------------------*/
if (RunType == 0 || RunType == 2)
{
printf("\nEnter drop standard deviations [-30 to 30; %d]: ", lastdrop);
gets(ctemp);
DropScore = atoi(ctemp);
if (DropScore < -30 || DropScore > 30) DropScore = lastdrop;
lastdrop = DropScore;
}
printf("\nUsing a drop score of %d standard deviations or %d.",
(int) DropScore/10, SMatrix->highpass);
/*--------- Re-score blocks and compute mean and sd -----------*/
totb = maxdups = 0;
fmean = fsdev = 0.0;
for (b=0; b<Total_Motifs; b++)
{
if (blocks[b].nmotif <= 0)
blocks[b].dropped = YES;
else
{
score_block(&blocks[b], seqs);
fmean += (double) blocks[b].t_score;
fsdev += (double) blocks[b].t_score*blocks[b].t_score;
totb++;
}
} /* end of for b */
/*-------- Compute mean and sd ----------------*/
if (totb > 1)
{
fsdev = fsdev - (double) fmean*fmean/totb;
fsdev = (double) fsdev/(totb-1);
if (fsdev > 0.0)
fsdev = sqrt(fsdev); /* standard deviation */
else fsdev = 0.0;
fmean = (double) fmean/totb; /* mean */
}
else
fsdev = 0.0;
printf("\nMean score=%.2f, standard deviation=%.2f", fmean, fsdev);
/*-----If sdev==0.0, then either all blocks have same score, or there
is only one block; don't want to drop them all ------------------*/
if (fsdev > 0.01) ftemp = fmean + fsdev * DropScore / 10;
else ftemp = 1.0;
TopScore = (long) ftemp - 1.0; /* Round down */
/* Possibly raise the drop score, but be careful not to drop
all the blocks! */
if (fmean >= SMatrix->highpass &&
TopScore < SMatrix->highpass) TopScore = SMatrix->highpass;
if (!AsIs)
{
printf("\nDropping blocks with score below %ld...", TopScore);
printf("\n unless they have more than one merged motif");
}
totb = 0;
for (b=0; b<Total_Motifs; b++)
if (!AsIs &&
blocks[b].dropped == NO &&
blocks[b].nmotif == 1 &&
blocks[b].t_score < (int) TopScore)
{
blocks[b].dropped = YES;
printf("\nBlock %d has been dropped for low score:", b);
print_head(&blocks[b]);
}
else if (blocks[b].dropped == NO)
{
totb++;
if (blocks[b].dups >maxdups) maxdups = blocks[b].dups;
cluster_seqs(&blocks[b], seqs);
printf("\nBlock %d:", b);
print_block(&blocks[b], seqs);
}
/* if (totb > MAXBLK)
totb = prune_blocks(blocks);
*/
printf("\nSurviving blocks=%d, max dups=%d", totb, maxdups);
/*------------------------------------------------------------------*/
/* Check each pair of surviving
blocks and look at those that overlap. By this point,
they don't overlap consistently in all the sequences or
they would have been merged. However, if they overlap
consistently in most of the sequences and if that overlap is
small, then trim one or both blocks so they don't overlap
any more. (Addresses the PS00031 problem without adding
gaps). See motomat2.c lines 140-157.
*/
Overlap = 0;
if (!AsIs)
{
printf("\nReducing block overlaps ...\n");
reduce_overlap(blocks, seqs);
}
/*---- Reduce significance level if there are dups. However, be
sure that RSignif > (NumSeqs-RSignif) to prevent cycles in the DAG ---*/
RSignif = Signif - ((int) maxdups/4);
if (RSignif <= (NumSeqs-RSignif) )
RSignif = ((int) NumSeqs/2) + 1; /* R must be more than half N */
if (RSignif < 2) RSignif = Signif; /* not sure what to do here */
if (RSignif < Signif)
printf(", reduced significance=%d", RSignif);
if (totb > 0) /* Don't try if no blocks left */
{
printf("\n\nLooking for best path...");
order_blocks(blocks);
build_dag(blocks);
paths = makepath();
find_paths(paths, blocks);
printf(", npath=%d", paths->nblocks);
if (paths->next_path != NULL)
{
printf(", nseqs=%d\n", paths->next_path->nseqs);
if (maxdups > 0) check_seqs(paths->next_path, blocks);
print_path(paths->next_path);
print_best(paths->next_path, blocks, seqs);
}
}
else
printf("\nNo surviving blocks");
fclose(Fblk);
printf("\n");
/*----------Print a summary of results in stderr--------------------*/
fprintf(stderr, "%s %d %d %d %d %d %d %d %d %d %d %d %d %ld %d %d %d\n",
ACName, NumSeqs, Signif, Dups, Distance, Total_Motifs, MinScore,
ClThres, DropScore, totb, paths->nblocks, paths->next_path->nbest,
paths->next_path->totmotif, paths->next_path->totscore,
paths->next_path->naas, paths->next_path->nseqs, Overlap);
fclose(err);
exit(0);
} /* end of main */
/*======================================================================
Print the motifs and sequences
========================================================================*/
void dump_data(motif, seqs)
struct motif_struct *motif;
struct sequences *seqs;
{
int i, j, k, offset, jj, kk;
printf("\nNumSeqs=%d, Total_Motifs=%d", NumSeqs, Total_Motifs);
k=0;
for (i=0; i<NumSeqs; i++)
{
printf("\nSequence #%d: %20s %d",
i, seqs->name+SNAMELEN*i, *(seqs->len+i) );
if (i>0) k += *(seqs->len+i-1);
printf(" %.5s", seqs->seq+k);
}
/*------NOTE: motif[].domain is the ACTUAL domain width, while in
blocks[].domain it is the displacement, or ACTUAL-1 -----*/
for (i=0; i<Total_Motifs; i++)
{
printf("\n\nMOTIF %d: %d(%d)%d(%d)%d", i, motif[i].aa1,
motif[i].distance1, motif[i].aa2, motif[i].distance2, motif[i].aa3);
printf(" domain=%d, score=%d, freq=%d, dups=%d, mots=%d",
motif[i].domain, motif[i].score, motif[i].freq, motif[i].dups,
motif[i].mots);
offset = (int) (motif[i].domain-motif[i].distance1-motif[i].distance2-1)/2;
/*>>>>>> Confused about this!
offset++;
>>>>>>>>*/
for (j=0; j<NumSeqs+motif[i].dups; j++)
{
jj = motif[i].seq_no[j];
kk = motif[i].pos[j] - offset;
printf("\n%20s (% 5d) ", seqs->name+SNAMELEN*jj, kk);
for (k=0; k<motif[i].domain; k++)
{
kk = motif[i].pos[j] + k - offset;
if (kk >= 0 && kk < *(seqs->len+jj) )
printf("%.1s", seqs->seq+seqs->offlen[jj]+kk);
else printf(".");
}
printf("\n ");
for (k=0; k<motif[i].domain; k++)
if (k == offset) printf("[");
else if (k == offset+motif[i].distance1) printf("|");
else if (k == offset+motif[i].distance1+motif[i].distance2)
printf("]");
else printf("_");
}
}
} /* end of dump_data */
/*=======================================================================*/
/* init_mm creates an array of merged motifs, each consisting of one */
/* motif, from an array of motifs created by H. Smith's program. */
/*=======================================================================*/
struct merged_motif *init_mm(motif)
struct motif_struct *motif;
{
struct merged_motif *new;
int m, s, ms;
double ftemp;
new = (struct merged_motif *) malloc(Total_Motifs * sizeof(struct merged_motif));
if (new == NULL)
{
printf("\nmerged_motif: Unable to allocate merged_motif structure!\n");
exit(-1);
}
for (m=0; m<Total_Motifs; m++)
{
new[m].dropped = NO;
new[m].aa[0] =(motif[m].aa1);
new[m].aa[1] =(motif[m].aa2);
new[m].aa[2] =(motif[m].aa3);
/* award a bonus if some motifs were merged in MOTIFJ */
/* if (motif[m].mots > 1) new[m].nmotif = 2;
*/
new[m].nmotif = 1;
new[m].nmotif = motif[m].mots;
new[m].nident = 0;
ftemp = (double) motif[m].domain;
new[m].max_score = (int) (motif[m].score/sqrt(ftemp));
/*--NOTE: block domain is displacement, not actual with ---*/
new[m].domain = new[m].t_domain = motif[m].domain - 1;
new[m].distance = motif[m].distance1 + motif[m].distance2;
new[m].dups = motif[m].dups;
/*--- loffset is not really offset, but 1st position ???--*/
/* NOTE: If the block width is the motif width, loffset is zero */
new[m].loffset = new[m].t_loffset =
(int) (motif[m].domain-motif[m].distance1-motif[m].distance2-1)/2;
new[m].t_score = -1; /* -1 => block has not been trimmed */
for (s=0; s<NumSeqs; s++)
{
new[m].leftpos[s] = -1;
new[m].cluster[s] = -1;
new[m].position[s] = -1;
}
new[m].maxpos = new[m].minpos = -1;
new[m].in_degree = new[m].out_degree = -1;
for (ms=0; ms<NumSeqs+motif[m].dups; ms++)
{
s = motif[m].seq_no[ms];
if (new[m].leftpos[s] < 0) /* Just take the 1st occurence */
new[m].leftpos[s] = motif[m].pos[ms];
}
for (ms=0; ms<motif[m].domain; ms++)
new[m].scores[ms] = motif[m].scores[ms];
};
return (new);
} /* end of init_mm */
/*=====================================================================
merge_blocks merges blocks if 1) ALL sequences are aligned the
same in all blocks and 2) the blocks overlap. Flag=0 means
check overlap in motif region, flag=1 means check overlap anywhere
in extended region.
NOTE: .t_loffset & .t_domain are used instead of .loffset and
.domain so the code will work for both raw (flag==0) and
extended (flag=1) blocks - since .t_loffset is initialized
to .loffset and .t_domain to .domain.
If the total width of the merged blocks would exceed
MAX_MERGE_WIDTH they are not merged. The amount of their overlap
is divided between them to make two contiguous, non-overlapping
blocks.
========================================================================*/
void merge_blocks(blocks, flag)
struct merged_motif *blocks;
int flag;
{
int b1, b2, s, right, left, ldiff, rdiff, lb1, lb2, rb1, rb2;
int overlap, ov1, ov2, bleft, bright, temp;
for (b1=0; b1<Total_Motifs-1; b1++)
if (blocks[b1].nmotif > 0)
for (b2=b1+1; b2 <Total_Motifs; b2++)
{
if (blocks[b2].nmotif > 0 &&
check_align(&blocks[b1], &blocks[b2]) &&
check_overlap(&blocks[b1], &blocks[b2], flag) )
{
/*--- lb1,2 and rb1,2 are the end positions of blocks b1 and b2
respectively for sequence 0. They are used to compute the relative
positions of the two blocks. This technique only works if 1) All
sequences are aligned the same way in both blocks, and 2) sequence
0 extends all the way across both blocks (?) ----*/
lb1 = blocks[b1].leftpos[0]-blocks[b1].t_loffset;
lb2 = blocks[b2].leftpos[0]-blocks[b2].t_loffset;
rb1 = lb1 + blocks[b1].t_domain;
rb2 = lb2 + blocks[b2].t_domain;
ldiff = lb1 - lb2; /*-- difference at left end*/
rdiff = rb1 - rb2; /*-- difference at right end*/
left = UMIN(lb1, lb2);
right = UMAX(rb1, rb2);
/*-------- Merge only if the total width is not too big ----------*/
if ( (right-left) < MAX_MERGE_WIDTH )
{
/*---- Update the block information ---*/
printf("\nMerging motif %d into motif %d", b2, b1);
blocks[b1].nmotif = blocks[b1].nmotif + 1;
blocks[b2].nmotif = 0;
if (blocks[b2].dups > blocks[b1].dups)
blocks[b1].dups = blocks[b2].dups;
if (!flag) blocks[b1].domain = right - left;
blocks[b1].t_domain = right - left;
left = UMIN(blocks[b1].leftpos[0], blocks[b2].leftpos[0]);
right = UMAX(blocks[b1].leftpos[0]+blocks[b1].distance,
blocks[b2].leftpos[0]+blocks[b2].distance);
blocks[b1].distance = right - left;
blocks[b1].t_loffset = left - UMIN(lb1, lb2);
if (!flag) blocks[b1].loffset = left - UMIN(lb1, lb2);
for (s=0; s<NumSeqs; s++)
blocks[b1].leftpos[s] =
UMIN(blocks[b1].leftpos[s], blocks[b2].leftpos[s]);
}
/*----------------------------------------------------------------------*/
/*--- Instead of merging, divide the blocks into non-overlapping blocks */
else if (flag)
{
bleft = bright = -1;
overlap = temp = 0;
if (ldiff < 0 && rdiff < 0) /* b1 is leftmost */
{
overlap = 1 + rb1 - lb2;
bleft = b1; bright = b2;
}
else if (ldiff > 0 && rdiff > 0) /* b2 is leftmost */
{
overlap = 1 + rb2 - lb1;
bleft = b2; bright = b1;
}
if (overlap > 0 && bleft >= 0 && bright >= 0)
{
ov1 = (int) (overlap/2); ov2 = overlap - ov1;
blocks[bleft].domain = blocks[bleft].domain - ov1;
if ( (blocks[bleft].loffset + blocks[bleft].distance) >
blocks[bleft].domain)
blocks[bleft].distance =
blocks[bleft].domain - blocks[bleft].loffset;
blocks[bleft].t_domain = blocks[bleft].t_domain - ov1;
blocks[bright].domain = blocks[bright].domain - ov2;
blocks[bright].t_domain = blocks[bright].t_domain - ov2;
blocks[bright].loffset = blocks[bright].loffset - ov2;
if (blocks[bright].loffset < 0)
{
temp = blocks[bright].loffset;
blocks[bright].distance = blocks[bright].distance
+ temp;
blocks[bright].loffset = 0;
}
blocks[bright].t_loffset = blocks[bright].t_loffset - ov2;
if (blocks[bright].t_loffset < 0)
blocks[bright].t_loffset = 0;
for (s=0; s<NumSeqs; s++)
blocks[bright].leftpos[s] =
blocks[bright].leftpos[s] - temp;
printf("\nBlocks %d and %d have been divided",
bleft, bright);
}
}
}
}
} /* end of merge_blocks */
/*=====================================================================
check_align returns YES if all the sequences in two blocks have the
same alignment.
========================================================================*/
int check_align(b1, b2)
struct merged_motif *b1, *b2;
{
int s, diff;
s=0;
diff = b1->leftpos[s]-b2->leftpos[s];
do {
s++;
} while (s<NumSeqs &&
b1->leftpos[s]-b2->leftpos[s] == diff);
if (s == NumSeqs) return(YES);
else return(NO);
} /* end of check_align */
/*=====================================================================
check_pos returns YES if all the sequences in two blocks have the
same left position.
========================================================================*/
int check_pos(b1, b2)
struct merged_motif *b1, *b2;
{
int s;
s=0;
/* Used to be
b1->leftpos[s] == b2->leftpos[s])
*/
while (s<NumSeqs &&
b1->leftpos[s] - b1->t_loffset == b2->leftpos[s] - b2->t_loffset)
s++;
if (s == NumSeqs) return(YES);
else return(NO);
} /* end of check_pos */
/*=====================================================================
check_overlap returns YES if the first sequence in each of two
blocks overlap anywhere within the blocks' motif area.
Flag=0 means check for overlap in motif region; flag=1 means check
anywhere in the extended region.
May not catch a block that overlaps in only some of the sequences,
but flag=2 forces it to check all sequences.
========================================================================*/
int check_overlap(b1, b2, flag)
struct merged_motif *b1, *b2;
int flag;
{
int left1, left2, s, over;
if (flag == 0) /* check motif regions */
{
/* Find start of 1st sequence in both blocks */
left1 = b1->leftpos[0];
left2 = b2->leftpos[0];
if ( (left2 >= left1 && left2 <= left1 + b1->distance) ||
(left1 >= left2 && left1 <= left2 + b2->distance) )
return(YES);
else return(NO);
}
else if (flag == 1) /* check full regions */
{
left1 = b1->leftpos[0] - b1->t_loffset;
left2 = b2->leftpos[0] - b2->t_loffset;
if ( (left2 >= left1 && left2 <= left1 + b1->t_domain) ||
(left1 >= left2 && left1 <= left2 + b2->t_domain) )
return(YES);
else return(NO);
}
else /* check ALL seqs in full regions */
{
s = 0; over = NO;
while (s<NumSeqs && !over)
{
left1 = b1->leftpos[s] - b1->t_loffset;
left2 = b2->leftpos[s] - b2->t_loffset;
if ( (left2 >= left1 && left2 <= left1 + b1->t_domain) ||
(left1 >= left2 && left1 <= left2 + b2->t_domain) )
over = YES;
s++;
}
return(over);
}
} /* end of check_overlap */
/*========================================================================
dup_blocks eliminates blocks that are exact duplicates by setting
the merged flag. Blocks are duplicates if they have the same
width & same offsets for all sequences
=========================================================================*/
void dup_blocks(blocks)
struct merged_motif *blocks;
{
int b1, b2;
for (b1=0; b1<Total_Motifs-1; b1++)
if (blocks[b1].nmotif > 0)
for (b2=b1+1; b2 <Total_Motifs; b2++)
{
if (blocks[b2].nmotif > 0 &&
blocks[b1].domain == blocks[b2].domain &&
check_pos(&blocks[b1], &blocks[b2]))
blocks[b2].nmotif = 0;
}
} /* end of dup_blocks */
/*========================================================================
reduce_overlap trims blocks that overlap only slightly but consistently
in most sequences
=========================================================================*/
void reduce_overlap(blocks, seqs)
struct merged_motif *blocks;
struct sequences *seqs;
{
int b1, b2, s, diff[MAXSEQS], bleft, bright, left, right, over[MAXSEQS];
int samediff, minover;
for (b1=0; b1<Total_Motifs-1; b1++)
{
if (blocks[b1].nmotif > 0)
{
for (b2=b1+1; b2 <Total_Motifs; b2++)
{
if ( blocks[b2].nmotif > 0 &&
check_overlap(&blocks[b1], &blocks[b2], 2) &&
!check_align(&blocks[b1], &blocks[b2]) )
{
/*>>> Find out how many overlap & by how much
if it's most & they all overlap by the
same amount, then trim the left one by reducing
->t_domain & re-score it */
samediff = YES;
minover = 0;
for (s=0; s<NumSeqs; s++)
{
diff[s] = blocks[b2].leftpos[s] - blocks[b1].leftpos[s];
if (s > 0)
{
if ( (diff[0] <= 0 && diff[s] > 0) ||
(diff[0] >= 0 && diff[s] < 0) ) samediff = NO;
}
if (diff[s] < 0) { bleft = b2; bright = b1; }
else { bleft = b1; bright = b2; }
left = blocks[bleft].leftpos[s] - blocks[bleft].t_loffset +
blocks[bleft].t_domain;
right = blocks[bright].leftpos[s] - blocks[bright].t_loffset;
over[s] = right - left - 1;
if (over[s] < minover) minover = over[s];
} /* end of sequence s */
/* Now check: 1) diff[s] same sign for all s
2) over[s] has a limited number of values, will be < 0
if there is an overlap: EG, if over = 0 for 2/10 seqs
and = -2 for 8/10, then trim 2 off the right end of bleft.
Count #different negative values in over[], if the
largest negative value is not too small, trim left block
by that amount. EG -10 is probably too much overlap,
but -5 may not be if left block is wide enough */
if (samediff && minover < 0 && minover >= -5)
{
/* Don't shorten if resulting block < 4 wide */
if ( (blocks[bleft].t_domain + minover) > MIN_DOMAIN_WIDTH )
{
blocks[bleft].t_domain += minover;
score_block(&blocks[bleft], seqs);
Overlap++;
printf("Block %d has been trimmed by %d to eliminate overlap with Block %d.\n",
bleft, minover, bright);
}
}
} /* end of if b1 and b2 overlap inconsistently */
} /* end of b2 */
} /* end of if b1 is still active */
} /* end of b1 */
} /* end of reduce_overlap */
/*=======================================================================
trim_block trims a block around the motifs.
It is mis-named, since what it actually does is EXTEND the block out
from the outer edges of the merged motifs up to the full width of
the block received from motifj. So it is the motifj block that is
trimmed.
=======================================================================*/
void trim_block(block)
struct merged_motif *block;
{
int left, right;
left = block->loffset;
do
{
left -= 1;
} while (left >= 0 && block->scores[left] >= MinScore);
right = block->loffset + block->distance;
do
{
right += 1;
} while (right <= block->domain &&
block->scores[right] >= MinScore);
/* update block only if it is wider than before */
if ( (right - left - 2) > block->t_domain)
{
block->t_loffset = block->loffset - left - 1;
block->t_domain = right - left - 2;
}
} /* end of trim_block */
/*=======================================================================*/
/* extend_block extends a block out from the motifs. */
/* Computes all possible blocks using all possible left and right
extensions. Takes block with maximum block score = Smith's score
divided by sqrt(width). */
/*=======================================================================*/
void extend_block(block)
struct merged_motif *block;
{
int left, middle, right, i, rn, ln, r, l, max_l, max_r;
double ftemp, motif_sum, max_score, l_score, r_score, l_sum, r_sum;
left = block->loffset; /* left edge of motif */
middle = block->distance + 1; /* width of motif */
right = left + block->distance; /* right edge */
motif_sum = 0;
for (i=left; i<=right; i++)
if (block->scores[i] >= SMatrix->highpass)
motif_sum += (block->scores[i]-SMatrix->highpass);
ftemp = (double) middle;
max_score = (double) ( (motif_sum/sqrt(ftemp)) );
max_l = left; max_r = right;
/*----------- Extend right only -----------------------------*/
r_sum = motif_sum; rn = middle;
for (r=right+1; r<=block->domain; r++)
{
rn++;
if (block->scores[r] >= SMatrix->highpass)
r_sum += (block->scores[r]-SMatrix->highpass);
ftemp = (double) rn;
r_score = (double) ( (r_sum/sqrt(ftemp)) );
if (r_score > max_score)
{
max_score = r_score; max_r = r; max_l = left;
}
}
/*--Extend left & try all possible right extensions for each left -----*/
l_sum = motif_sum; ln = middle;
for (l=left-1; l>=0; l--)
{
ln++;
if (block->scores[l] >= SMatrix->highpass)
l_sum += block->scores[l]-SMatrix->highpass;
ftemp = (double) ln;
l_score = (double) ( (l_sum/sqrt(ftemp)) );
if (l_score > max_score)
{
max_score = l_score; max_l = l; max_r = right;
}
r_sum = l_sum; rn = ln;
for (r=right+1; r<=block->domain; r++) /*--- all possible right ---*/
{
rn++;
if (block->scores[r] >= SMatrix->highpass)
r_sum += (block->scores[r]-SMatrix->highpass);
ftemp = (double) rn;
r_score = (double) ( (r_sum/sqrt(ftemp)) );
if (r_score > max_score)
{
max_score = r_score; max_r = r; max_l = l;
}
}
}
/*------Now max_r and max_l are the best right & left extensions ------- */
block->t_loffset = block->loffset - max_l;
block->t_domain = max_r - max_l;
block->t_score = max_score;
} /* end of extend_block */
/*=======================================================================
Print out a trimmed block.
=========================================================================*/
void print_block(block, seqs)
struct merged_motif *block;
struct sequences *seqs;
{
int s, i, j, x, sout;
char tempname[30];
print_head(block);
if (RunType==0 || RunType == 2 || Debug)
{
printf("\nScore ");
for (i= block->loffset-block->t_loffset;
i<=block->loffset-block->t_loffset+block->t_domain;
i++)
{
x = (int) (block->scores[i]/100);
if (x >= 10) strcpy (tempname, "*");
else kr_itoa(x, tempname, 10);
/* Print sequence characters at the motif extremes
Just takes 1st sequence value... */
if (i == block->loffset ||
i == block->loffset+block->distance)
{
j = block->leftpos[0] - block->loffset + i;
strncpy(tempname, seqs->seq+seqs->offlen[0]+j, 1);
}
printf("%1s", tempname);
}
/*---- Print out in cluster order -------------*/
sout = 0; /* # seqs printed for block */
for (s=0; s<NumSeqs; s++)
if (block->cluster[s] < 0) /* unclustered first */
{
print_seq(block, seqs, s);
sout++;
}
i = 0; /* cluster # */
while (sout < NumSeqs)
{
for (s=0; s<NumSeqs; s++)
if (block->cluster[s] == i)
{
print_seq(block, seqs, s);
sout++;
}
i++;
}
} /* end of RunType check */
} /* end of print_block */
/*======================================================================*/
void print_head(block)
struct merged_motif *block;
{
printf("\n %.1s%.1s%.1s (extended width=%d, score=%d",
num_to_aachar(block->aa[0]),
num_to_aachar(block->aa[1]),
num_to_aachar(block->aa[2]),
block->t_domain+1, block->t_score );
printf(" motifs=%d, motif width=%d, conserved=%d)",
block->nmotif, block->distance+1, block->nident);
} /* end of print_head */
/*=======================================================================
Save a block. Include the sequences flagged in pseqs.
Requires global variable FILE *Fblk
=========================================================================*/
void save_block(block, seqs, pseqs)
struct merged_motif *block;
struct sequences *seqs;
int pseqs[MAXSEQS]; /* seqs to print */
{
int s, i, sout, prevsout, pumseqs;
/*--------------- Build the output file name --------------------------*/
/* AC, MinDist and MaxDist are set in print_best() */
printf("\nSaving block %.1s%.1s%.1s",
num_to_aachar(block->aa[0]),
num_to_aachar(block->aa[1]),
num_to_aachar(block->aa[2]) );
/*-------------- Open and write the output file -----------------------*/
fprintf(Fblk, "ID %s; BLOCK\n", ID);
fprintf(Fblk, "AC %s;", AC);
fprintf(Fblk, " distance from previous block=(%d,%d)\n",
MinDist, MaxDist);
fprintf(Fblk, "DE %s\n", DE);
if (Gibbs)
fprintf(Fblk, "BL %.1s%.1s%.1s gibbs=[%d,%d,%d] motomat=[%d,%d,%d]",
num_to_aachar(block->aa[0]),
num_to_aachar(block->aa[1]),
num_to_aachar(block->aa[2]),
Signif, Dups, Distance,
MinScore, ClThres, DropScore);
else
fprintf(Fblk, "BL %.1s%.1s%.1s motif=[%d,%d,%d] motomat=[%d,%d,%d]",
num_to_aachar(block->aa[0]),
num_to_aachar(block->aa[1]),
num_to_aachar(block->aa[2]),
Signif, Dups, Distance,
MinScore, ClThres, DropScore);
fprintf(Fblk, " width=%d", block->t_domain+1);
/*-------------- Write sequences in cluster order --------------------*/
pumseqs = 0; /* # seqs to be printed for block */
for (s=0; s<NumSeqs; s++)
if (pseqs[s] == YES) pumseqs += 1;
fprintf(Fblk, " seqs=%d\n", pumseqs);
sout = 0; /* # seqs printed for block */
for (s=0; s<NumSeqs; s++)
if (pseqs[s] == YES && block->cluster[s] < 0) /* unclustered first */
{
write_seq(Fblk, block, seqs, s);
fprintf(Fblk, "\n\n");
sout++;
}
i = 0; /* cluster # */
while (sout < pumseqs)
{
prevsout = sout;
for (s=0; s<NumSeqs; s++)
if (pseqs[s] == YES && block->cluster[s] == i)
{
write_seq(Fblk, block, seqs, s);
fprintf(Fblk, "\n");
sout++;
}
if (sout != pumseqs && sout > prevsout)
fprintf(Fblk, "\n"); /* Extra line bw clusters*/
i++;
}
fprintf(Fblk, "//\n");
} /* end of save_block */
/*======================================================================*/
/* Cluster sequences in a trimmed block based on the number of */
/* identities within the block. */
/* 1. Compute number of identities for each possible pair of seqs. */
/* Results stored in lower half of matrix (pairs). */
/* 2. Use clustering threshold % of # of AAs in trimmed block. */
/* 3. Cluster recursively by traversing cols, rows of matrix. */
/*======================================================================*/
void cluster_seqs(block, seqs)
struct merged_motif *block;
struct sequences *seqs;
{
int nclus, npair, threshold, s1, s2, l1, l2, px, i, i1, i2;
int oldclus, minclus;
struct pair *pairs;
npair = (NumSeqs*(NumSeqs-1))/2;
pairs = (struct pair *) malloc (npair * sizeof(struct pair));
if (pairs == NULL)
{
printf("\ncluster_seqs: Unable to allocate pair structure!\n");
exit(-1);
}
threshold = (int) (ClThres*(block->t_domain+1))/100;
/* Compute scores for all possible pairs of sequences */
for (s1=0; s1<NumSeqs-1; s1++) /* col = 0, n-2 */
{
l1 = block->leftpos[s1] - block->t_loffset;
for (s2=s1+1; s2<NumSeqs; s2++) /* row = col+1, n-1 */
{
l2 = block->leftpos[s2] - block->t_loffset;
px = INDEXCOLROW(NumSeqs, s1, s2);
pairs[px].score = 0;
pairs[px].cluster = -1;
for (i=0; i<=block->t_domain; i++)
{
i1 = l1+i; i2 = l2+i;
if (i1 >= 0 && i1 < *(seqs->len+s1) &&
i2 >= 0 && i2 < *(seqs->len+s2) &&
strncmp(seqs->seq + seqs->offlen[s1] + i1,
seqs->seq + seqs->offlen[s2] + i2, 1) == 0 )
pairs[px].score += 1;
}
} /* end of s2 */
} /* end of s1 */
/* Print scores */
/* printf("\nThreshold=%d", threshold);
for (s2=1; s2<NumSeqs; s2++)
{
printf ("\n");
for (s1=0; s1<s2; s1++)
{
px = INDEXCOLROW(NumSeqs, s1, s2);
printf(" %.3d", pairs[px].score);
}
}
*/
/*-------Cluster if score exceeds threshold by scanning cols (s1) */
for (s1=0; s1<NumSeqs; s1++)
block->cluster[s1] = -1; /* clear out old values */
nclus = 0;
for (s1=0; s1<NumSeqs-1; s1++) /* col = 0, n-2 */
for (s2=s1+1; s2<NumSeqs; s2++) /* row = col+1, n-1 */
{
px = INDEXCOLROW(NumSeqs, s1, s2);
if (pairs[px].score >= threshold)
{
if (block->cluster[s1] < 0) /* s1 has no cluster yet */
{
if (block->cluster[s2] < 0) /* new cluster */
{
block->cluster[s1] = nclus++;
block->cluster[s2] = block->cluster[s1];
}
else /* s2 has a cluster, use it */
block->cluster[s1] = block->cluster[s2];
}
/* use s1's cluster if it has one and s2 doesn't */
else if (block->cluster[s1] >= 0 && block->cluster[s2] < 0)
block->cluster[s2] = block->cluster[s1];
/* merge the two clusters into the lower number */
else if (block->cluster[s1] >= 0 && block->cluster[s2] >= 0)
{
minclus = block->cluster[s1]; oldclus = block->cluster[s2];
if (block->cluster[s2] < block->cluster[s1])
{ minclus = block->cluster[s2]; oldclus = block->cluster[s1]; }
for (i1=0; i1<NumSeqs; i1++)
if (block->cluster[i1] == oldclus)
block->cluster[i1] = minclus;
}
}
} /* end of s2 */
free(pairs);
} /* end of cluster_seqs */
/*======================================================================*/
/* The blocks score is the sum of the column scores at or over HighPass
divided by the square root of the block width.
Scores only trimmed blocks. */
/*======================================================================*/
void score_block(block, seqs)
struct merged_motif *block;
struct sequences *seqs;
{
int a, s, col, aa[20], pos, offset, flag;
double ftemp, block_sum, ident_sum, weights[MAXSEQS];
char c1, c[2];
pb_weights(block, seqs, weights); /* add sequence weights */
block->nident = 0; /* could be a re-count */
block->t_score = 0;
block_sum = ident_sum = 0.0;
offset = block->loffset - block->t_loffset;
for (col=0; col<=block->t_domain; col++)
{
flag = NO; /* flag this col if an ident */
for (a=0; a<20; a++) aa[a] = 0;
for (s=0; s<NumSeqs; s++)
{
pos = block->leftpos[s] - block->t_loffset + col;
if (pos >= 0 && pos < *(seqs->len+s) )
{
strncpy(c, seqs->seq + seqs->offlen[s] + pos, 1);
c1 = c[0];
a = aachar_to_num(c1);
aa[a]++;
}
}
/* Note: may be more than one aa in a col at least Signif times,
but are only counting such a column once */
for (a=0; a<20; a++) if (aa[a] >= Signif) flag = YES;
if (flag)
{
block->nident += 1;
ident_sum += (block->scores[col+offset] - SMatrix->highpass);
}
if (block->scores[col+offset] >= SMatrix->highpass)
block_sum += (block->scores[col+offset] - SMatrix->highpass);
} /* end of col */
/*printf(" block_sum=%d, ident_sum=%d", (int) block_sum, (int) ident_sum); */
/* ftemp = (double) block->t_domain + 1 + block->nident; */
ftemp = (double) block->t_domain + 1;
if (ftemp > 0)
block->t_score = (int) ( block_sum/sqrt(ftemp) );
} /* end of score_block */
/*======================================================================
Put position-based sequence weights in the weights array
col = block column, pos = sequence position
======================================================================*/
void pb_weights(block, seqs, weights)
struct merged_motif *block;
struct sequences *seqs;
double weights[MAXSEQS];
{
struct pb_counts *pb;
double factor, dtemp;
int seq, pos, col, aa, width;
char c1, c[2];
width = block->t_domain;
pb = (struct pb_counts *) malloc(width*sizeof(struct pb_counts));
for (col = 0; col < width; col++)
{
pb[col].diffaas = 0.0;
for (aa = 0; aa < 20; aa++)
pb[col].naas[aa] = (double) 0.0;
}
for (col = 0; col < width; col++)
for (seq = 0; seq < NumSeqs; seq++)
{
pos = block->leftpos[seq] - block->t_loffset + col;
if (pos >= 0 && pos < *(seqs->len+seq) )
{
strncpy(c, seqs->seq + seqs->offlen[seq] + pos, 1);
c1 = c[0];
aa = aachar_to_num(c1);
if (aa >= 0 && aa <= 19)
{
pb[col].naas[aa] += 1;
}
/*
else
{
printf("pb_weights:%d ignored\n", aa);
}
*/
}
}
factor = 1.0;
for (col = 0; col < width; col++)
{
for (aa = 0; aa < 20; aa++)
{
if (pb[col].naas[aa] > 0.0)
{
pb[col].diffaas += 1; /* # of different types of aas in col */
}
}
}
for (seq = 0; seq < NumSeqs; seq++)
{
weights[seq] = 0.0;
for (col = 0; col < width; col++)
{
pos = block->leftpos[seq] - block->t_loffset + col;
if (pos >= 0 && pos < *(seqs->len+seq) )
{
strncpy(c, seqs->seq + seqs->offlen[seq] + pos, 1);
c1 = c[0];
aa = aachar_to_num(c1);
dtemp = pb[col].diffaas * pb[col].naas[aa];
if (dtemp > 0.0) weights[seq] += 1.0 / dtemp;
}
}
}
/* Normalize weights to add to NumSeqs */
dtemp = 0.0;
for (seq = 0; seq < NumSeqs; seq++) dtemp += weights[seq];
for (seq = 0; seq < NumSeqs; seq++)
weights[seq] *= ((double) NumSeqs/dtemp);
free(pb);
} /* end of pb_weights */
/*======================================================================*/
/* Print a sequence */
/*======================================================================*/
void print_seq(block, seqs, s)
struct merged_motif *block;
struct sequences *seqs;
int s;
{
int left, right, i;
left = block->leftpos[s] - block->t_loffset;
right = left + block->t_domain;
printf("\n%20s (% 5d) ", seqs->name+SNAMELEN*s, left+1);
for (i=left; i<=right; i++)
{
if (i >= 0 && i < *(seqs->len+s) )
printf("%.1s", seqs->seq+seqs->offlen[s]+i );
else
printf(".");
}
printf(" % 3d", block->cluster[s]);
} /* end of print_seq */
/*======================================================================*/
/* Write a sequence to an output file */
/*======================================================================*/
void write_seq(out, block, seqs, s)
FILE *out;
struct merged_motif *block;
struct sequences *seqs;
int s;
{
int left, right, i;
left = block->leftpos[s] - block->t_loffset;
right = left + block->t_domain;
/* Print out the actual starting position of the sequence in
the block, not the offset to it */
fprintf(out, "%20s (% 5d) ", seqs->name+SNAMELEN*s, left+1);
for (i=left; i<=right; i++)
{
if (i >= 0 && i < *(seqs->len+s) )
fprintf(out, "%.1s", seqs->seq+seqs->offlen[s]+i );
else
fprintf(out, "X");
}
} /* end of write_seq */
/*======================================================================*/
/* Calculate column scores for a block.
Looks at all possible pairs of sequence AAs in a block column
and takes average score. If one or both sequences has no AA in
a column, because the block goes off one end or the other of the
sequence, uses the score for 'J' */
/*======================================================================*/
void score_cols(block, seqs)
struct merged_motif *block;
struct sequences *seqs;
{
int col, s1, s2, sum, npair, col1, col2;
double weights[MAXSEQS];
char c1, c2, c[2];
pb_weights(block, seqs, weights); /* add sequence weights */
npair = (int) NumSeqs * (NumSeqs-1) / 2; /* all possible pairs */
for (col=0; col<=block->domain; col++)
{
sum = 0; /* sum of all AA pairs */
for (s1=0; s1<NumSeqs; s1++)
{
col1 = block->leftpos[s1] - block->loffset + col;
if (col1 >= 0 && col1 < *(seqs->len+s1) )
{
strncpy(c, seqs->seq+seqs->offlen[s1]+col1, 1);
c1 = c[0];
}
else { c1 = 'X';} /* off end of seq 1 */
for (s2=s1+1; s2<NumSeqs; s2++)
{
col2 = block->leftpos[s2] - block->loffset + col;
if (col2 >= 0 && col2 < *(seqs->len+s2) )
{
strncpy(c, seqs->seq+seqs->offlen[s2]+col2, 1);
c2 = c[0];
sum += SMatrix->scores[aachar_to_num(c1)][aachar_to_num(c2)];
}
else { c2 = 'X';} /* off end of seq 2 */
} /* end of s2 */
} /* end of s1 */
block->scores[col] = (int) ((long int) sum * 100 / npair);
} /* end of col */
} /* end of score_cols */
/*======================================================================*/
/* Increase drop score and restart motomat. */
/*======================================================================*/
void restart()
{
char arg2[5], arg3[5], arg4[5];
int row, col;
printf("\n>>>MOTOMAT out of memory!");
if (DropScore <= 28)
{
DropScore += 2;
printf(" Restarting with DropScore = %d\n", DropScore);
kr_itoa(MinScore, arg2, 10);
kr_itoa(ClThres, arg3, 10);
kr_itoa(DropScore, arg4, 10);
/*--- Have to free enough memory to reload motomat ----------------------*/
for (row=0; row<Total_Motifs; row++)
for (col=0; col<Total_Motifs; col++)
if (Dag[row][col] != NULL) free(Dag[row][col]);
execlp(Argv0, Argv0, Mot_Filename, arg2, arg3, arg4, NULL);
exit(0);
}
else exit(0);
} /* end of restart */
/*======================================================================*/
#include "motomat2.c"
|