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 1651 1652 1653 1654 1655 1656 1657 1658 1659 1660 1661 1662 1663 1664 1665 1666 1667 1668 1669 1670 1671 1672 1673 1674 1675 1676 1677 1678 1679 1680 1681 1682 1683 1684 1685 1686 1687 1688 1689 1690 1691 1692 1693 1694 1695 1696
|
/*=====================================================================*/
/*(C) Copyright 1991-9 by Fred Hutchinson Cancer Research Center */
/* multimat.c reads several BLIMPS MATRIX search output files
and merges the results. Writes stats to multimat.dat.
eg: multimat 10 blocks.dat PS00094.lis PS00094*.hom or
multimat 15 none none PS00094A.hom PS00094B.hom ... (up to 26)
1st arg = number of hits to report
2nd arg = name of blocks database (use non-existent file name
to skip it, if file is found, will open it & compare
search results with blocks of same name)
3rd arg = name of list of true positive sequences (use
non-existent file name to omit it) If file is found
will only report hits NOT in file.
4rd arg = name of matrix search output files, etc.
EG PS00094A, PS00094B, PS00094C
Name restrictions:
AC name in the blocks in the blocks database (2nd arg)
must be unique and <= 8 characters long.
The blocks database must be sorted in AC order.
It is assumed that all of the blocks from the
same family have all the same AC except for the
last character.
Search result file names (4th arg) must match the AC names of
the blocks that were the queries for the search.
Sequence names in the blocks in the blocks db (2nd arg)
and in the list of true positives (3rd arg)
must be unique, and must match the sequence
names in the database searched. It is assumed that
searches are for blocks belonging to the same
family.
--------------------------------------------------------------------------
8/1/91 J. Henikoff
5/15/99 1. Format changes for Blimps 3.2.6
6/26/99 1. Sequence name length from 10 to 18
6/11/00 1. AC length from 7 to 9
12/23/06 1. Sequence name length from 18 to 20
====================================================================*/
#include "motifj.h"
#define SWISS 69113 /* Swiss35 */
#define NSCORE 5000 /* Maximum # of scores in .hom file */
#define MAXHOM 26 /* Maximum # of .hom files */
#define MAXHIT 10 /* Maximum # of hits to report */
#define MAXMEM 64000 /* Maximum bytes for DOS array */
#define SEEK_SET 0
#define PROTEIN 1
#define DNA 3
#define MAXMAP 60 /* Maximum map width in characters */
#define MAX_WIDTH 100 /* Maximum block width expected */
struct hom { /* search results structure */
char ac[MAXAC+1]; /* block name, eg. PS00094A */
int norm; /* normalized score */
int max_norm; /* max normalized score for this group */
int min_rank; /* minimum rank for this group of blocks */
int rank; /* rank of this result in search */
int frame; /* frame of alignment */
int strength; /* block strengh */
int score; /* patmat score */
int offset; /* offset of alignment */
double seqlen; /* db sequence length */
int width; /* width of block */
char title[30]; /* block description */
char seq_id[25]; /* patmat sequence id */
char aa[MAX_WIDTH]; /* alignment to block */
int map_flag; /* used by map_blocks */
int tp; /* true positive flag, 1 if tp, 2 if in block*/
};
struct best_block { /* block structure */
char ac[MAXAC+1]; /* accession number */
int nseq, width, strength; /* #seqs, width, strength */
int s995; /* 99.5 percentile score */
int minprev, maxprev; /* distances from prev block */
char name[MAXSEQS][SNAMELEN+5]; /* name of seq */
int offset[MAXSEQS]; /* offset of seq */
char aa[MAXSEQS][MAX_WIDTH]; /* aas for seq */
int cluster[MAXSEQS]; /* cluster # for seq */
int ncluster[MAXSEQS]; /* #seqs in same cluster */
long dat_pos; /* position of block in blocks.dat*/
int query; /* score if last seq is a query */
int rank; /* rank if last seq is a query */
int anchor; /* anchor block flag */
struct best_block *next_block; /* next block in path */
};
/*---- Functions also in blksort.c -----*/
void print_blurb();
void check_dat();
struct best_block *read_block();
void fill_block();
int add_query();
double hypergeo();
int distance_okay();
int prev_dist();
int distance();
void map_blocks();
void align_blocks();
int closest_seq();
int tempcmp();
int read_hom();
int check_tp();
void show_hom();
int id_cmp();
int rank_cmp();
int idnorm_cmp();
int norm_cmp();
struct best_block *get_blocks();
void consensus();
int compute_loc();
void re_normalize();
/*--------------------Routines from motmisc.o------------------------*/
struct db_id *makedbid();
int get_ids();
struct split_name *split_names();
void kr_itoa();
/*------------Global variables -------------*/
char Version[12] = " 6/11/00.1";/* Version number */
int Ask; /* Interactive flag */
char AC[30]; /* Global variables */
int ACLen; /* Length of AC */
int NScore; /* Max# scores for DOS */
int LisSeq; /* #seqs in TP list file */
int FragSeq;
int BlkSeq;
int SeqType; /* DNA or PROTEIN database searched */
char Query[SNAMELEN]; /* Query file name */
double DBLen; /* DB sequence length */
int NBlock; /* # blocks in family */
double NSeq; /* # sequences searched */
double NRank; /* # of ranks assigned in search */
int DBType; /* =1 if amino acid, =3 if nucleotide */
int MaxHit; /* # hits to report */
FILE *Fout; /* multimat.lis statistics file (not TPs) */
FILE *Fdat; /* multimat.dat statistics file (TPs) */
FILE *Fmis; /* multimat.mis file (missed TPs) */
FILE *Ffnd; /* multimat.fnd file (missed TPs) */
/*======================================================================*/
int main(argc, argv)
int argc;
char *argv[];
{
FILE *fhom, *fblk, *flis;
char homfile[MAXHOM][FNAMELEN], *ptr, *ptr1, ctemp[FNAMELEN];
char datfile[FNAMELEN], lisfile[FNAMELEN];
int totscores, i, nhom;
struct hom *temp;
struct db_id *ids, *did;
struct best_block *blocks;
printf("MULTIMAT: (C) Copyright 1991 by Fred Hutchinson Cancer ");
printf("Research Center\n");
printf("Version %s\n", Version);
if (argc <= 4)
{
printf("USAGE: multimat <nhits> <blocks> <list> <blimps>\n");
printf(" <nhits> = Maximum # of hits to report\n");
printf(" <blocks> = File containing blocks for comparison\n");
printf(" <list> = File containing list of TP sequences\n");
printf(" <blimps> = Blimps search result files, ");
printf(" blocks vs sequence db\n");
}
/* Fout = fopen("multimat.lis", "a"); */
Fdat = fopen("multimat.dat", "a"); /* statistics file */
/*--------------arg 1: Number of hits to report---------------------*/
MaxHit = MAXHIT;
ctemp[0] = '\0';
if (argc > 1) strcpy(ctemp, argv[1]);
else
{
printf("\nEnter maximum number of hits to report [%d]: ", MaxHit);
gets(ctemp);
}
if (strlen(ctemp)) MaxHit = atoi(ctemp);
if (MaxHit < 1 || MaxHit > NSCORE) MaxHit = MAXHIT;
/*------------- arg 2: blocks database file -----------------------------*/
if (argc > 2)
strcpy(datfile, argv[2]);
else
{
printf("\nEnter name of blocks database:\n");
gets(datfile);
}
if ( (fblk=fopen(datfile, "r")) == NULL)
{
printf("\nCannot open file %s", datfile);
printf(";\nsearch results will not be compared with blocks database.");
}
else
{
printf("\nSearch results will be compared with %s", datfile);
}
/*------------- arg 3: .lis file ----------------------------------------*/
if (argc > 3)
strcpy(lisfile, argv[3]);
else
{
printf("\nEnter name of file containing list of true positives:\n");
gets(lisfile);
}
if ( (flis=fopen(lisfile, "r")) == NULL)
{
printf("\nCannot open file %s;", lisfile);
printf("\nsearch results will not be compared with true positive list.\n");
LisSeq = FragSeq = BlkSeq = 0; ids = NULL;
}
else
{
printf("\nSearch results will be compared with %s;", lisfile);
printf("\nonly hits not in this list will be reported\n");
ids = makedbid();
printf("%d IDs in %s\n", (LisSeq=get_ids(flis, ids)), lisfile);
fclose(flis);
/*----------- Count the true positive sequences -------------------*/
LisSeq = FragSeq = BlkSeq = 0;
did = ids->next;
while (did != NULL)
{
did->found = NO;
if (!did->block && !did->frag) LisSeq++;
if (did->block) BlkSeq++;
if (did->frag) FragSeq++;
did = did->next;
}
Fmis = fopen("multimat.mis", "w"); /* statistics file */
Ffnd = fopen("multimat.fnd", "w"); /* statistics file */
}
/*------------- args 4-: .hom files -------------------------------*/
nhom = 0;
if (argc > MAXHOM+4) argc=MAXHOM+4;
if (argc > 4)
{
for (i=4; i<argc; i++)
strcpy(homfile[nhom++], argv[i]);
}
else
{
do
{
printf("\nEnter name of BLIMPS results file: ");
gets(homfile[nhom++]);
} while (nhom <= MAXHOM && strlen(homfile[nhom-1]));
nhom--;
}
if (nhom < 1)
{
printf("\nNo results files specified.\n");
exit(-1);
}
/*----------Create an array big enough for NSCORE scores per file------*/
NScore=NSCORE;
/*DOS For DOS, have to restrict max size of array to 64k
ltemp = MAXMEM; ltemp = ltemp/(nhom*sizeof(struct hom));
NScore = (int) ltemp;
DOS*/
temp = (struct hom *) malloc(nhom * NScore * sizeof(struct hom));
if (temp==NULL)
{
printf("\n\nNOT ENOUGH MEMORY!\n");
exit(-1);
}
/*-------------Now read the files into the array for sorting-----------*/
totscores = 0;
DBType = PROTEIN; /* BLIMPS always reports translated length */
for (i=0; i<nhom; i++)
{
if ( (fhom=fopen(homfile[i], "r")) == NULL)
{
printf("\nCannot open file %s\n", homfile[i]);
exit(-1);
}
else printf("\nReading %s...", homfile[i]);
ptr = strrchr(homfile[i],'/'); /* look for file name */
if (ptr != NULL) ptr1 = strtok(ptr+1, ".");
else ptr1 = strtok(homfile[i], ".");
strcpy(AC, ptr1);
NScore=read_hom(fhom, temp, totscores, ids);
printf("\n%d scores read from %s\n", NScore, homfile[i]);
totscores += NScore;
fclose(fhom);
}
/* AC is assumed to be a family name; all but the last character of
the file name, maximum of 9 characters */
if (nhom > 1) AC[strlen(AC)-1] = '\0';
if ((int) strlen(AC) > 9)
{
printf("\nWARNING: AC family name truncated to 9 characters. ");
AC[9] = '\0';
}
printf("AC family name = %s\n", AC);
ACLen = strlen(AC);
/*------------------Make a list of the blocks in the database----*/
/* For this AC => all but last character of AC must be the same for */
/* all block queries */
if (fblk != NULL)
{
blocks = get_blocks(fblk, AC);
re_normalize(temp, totscores, blocks);
fclose(fblk);
}
else blocks = NULL;
/*-----------------Sort and present the results -----------------------*/
show_hom(temp, totscores, blocks);
/*------------Write a file of the tps not found ------------------------*/
if (ids != NULL)
{
did = ids->next;
while (did != NULL)
{
if (!did->found)
{
fprintf(Fmis, "%s", did->entry);
if (did->block) fprintf(Fmis, "\tBLOCK");
if (did->frag) fprintf(Fmis, "\tFRAGMENT");
fprintf(Fmis, "\n");
}
else
{
fprintf(Ffnd, "%s", did->entry);
if (did->block) fprintf(Ffnd, "\tBLOCK");
if (did->frag) fprintf(Ffnd, "\tFRAGMENT");
fprintf(Ffnd, "\n");
}
did = did->next;
}
}
/*---------------------------------------------------------------------*/
if (Fout != NULL) fclose(Fout);
if (Fdat != NULL) fclose(Fdat);
if (Fmis != NULL) fclose(Fmis);
if (Ffnd != NULL) fclose(Ffnd);
printf("\n");
exit(0);
} /* end of main */
/*============================================================================
BLIMPS (BLocks IMProved Searcher) Version 3.2.5 1998/05
(C) Copyright 1993-2000, Fred Hutchinson Cancer Research Center
Block File: /howard/jorja/seqs/PACA.blk
Target File (s) : stuff.dna
Records Searched: 18
Scores Done: 108
Alignments Done: 32866
AC# Description Score RF AA# Length
pa01l_141.s1 0 CHROMAT_FILE: pa01l_141.s1 PHD_FILE: pa01l_141.s1.phd.1 TI 2081 -1 116 260 rYArlvKemsEkvqfiyithnkiaMEMAdqlmgvTmhEpgcsrlVavDveeav
NOTE: Uses hard-coded positions on Blimps output line, mainly because there
are spaces in the Description.
=============================================================================*/
int read_hom(fhom, temp, first, tps)
FILE *fhom;
struct hom *temp;
int first;
struct db_id *tps;
{
int t, i, v324, v326, minlen, offscore, offframe, offoffset, offlen;
int offtitle;
char line[MAXLINE], ctemp[30], *ptr;
SeqType = PROTEIN; /* default database type */
t = first; /* first available position in temp array */
v324 = NO; /* format changed with Blimps 3.2.4 */
v326 = YES; /* format changed again with Blimps 3.2.6 */
minlen = 105; offscore = 84; offframe = 89; offoffset = 92; offlen = 98;
offtitle = 22;
while (!feof(fhom) && fgets(line, MAXLINE, fhom) != NULL)
{
if (strstr(line, "Version 3.2.3") ||
strstr(line, "Version 3.2.2") || strstr(line, "Version 3.2.1") )
{
v324 = v326 = NO;
minlen = 96; offscore = 75; offframe = 80; offoffset = 83; offlen = 89;
offtitle = 14;
}
else if ( strstr(line, "Version 3.2.4") || strstr(line, "Version 3.2.5") )
{
v324 = YES; v326 = NO;
minlen =105; offscore = 84; offframe = 89; offoffset = 92; offlen = 98;
offtitle = 24;
}
else if (strstr(line, "Records Searched:"))
{
ptr = strtok(line, ":"); ptr = strtok(NULL, "\t\r\n");
NSeq = atof(ptr);
}
else if (strstr(line, "Scores Done:"))
{
ptr = strtok(line, ":"); ptr = strtok(NULL, "\t\r\n");
NRank = atof(ptr);
}
else if ((int) strlen(line) > minlen && (t-first) < NScore)
{
strcpy(temp[t].ac, AC);
if (v324 || v326)
{
strncpy(temp[t].seq_id, &line[0], 20); temp[t].seq_id[20]='\0';
}
else
{
strncpy(temp[t].seq_id, &line[0], 12); temp[t].seq_id[12]='\0';
}
strncpy(temp[t].title, &line[offtitle], 19);
temp[t].title[19] = '\0';
strncpy(ctemp, &line[offscore], 4); ctemp[4]='\0';
temp[t].score = atoi(ctemp);
strncpy(ctemp, &line[offframe], 2); ctemp[2]='\0';
temp[t].frame = atoi(ctemp);
if (temp[t].frame != 0) SeqType = DNA;
/* BLIMPS frames are only non-zero for DNA databases */
if (temp[t].frame != 0) DBType = DNA;
strncpy(ctemp, &line[offoffset], 5); ctemp[5]='\0';
temp[t].offset = atoi(ctemp);
strncpy(ctemp, &line[offlen], 6); ctemp[6]='\0';
temp[t].seqlen = atof(ctemp);
temp[t].width = 0; i=minlen;
/* truncates wide blocks */
while (line[i] != '\n' && temp[t].width < MAX_WIDTH)
{
temp[t].aa[temp[t].width] = line[i++];
temp[t].width += 1;
}
temp[t].aa[temp[t].width] = '\0';
strcpy(ctemp, temp[t].seq_id);
if (tps != NULL)
temp[t].tp = check_tp(ctemp, tps);
temp[t].min_rank = temp[t].rank = t-first;
temp[t].map_flag = NO;
t++;
}
}
/* Compute normalized score */
for (i = first; i < t; i++)
{
temp[i].norm = round((float) 1000. * temp[i].score / temp[t-1].score);
temp[i].max_norm = temp[i].norm;
}
return(t-first);
} /* end of read_hom */
/*=====================================================================
See if sequence id is in the list of true positives
Write info to multimat.dat if it's a TP not in the blocks
========================================================================*/
int check_tp(id, tps)
char *id;
struct db_id *tps;
{
struct db_id *did;
did = tps->next;
while (did != NULL)
{
if (strncmp(id, did->entry, strlen(did->entry)) == 0)
{
did->found = YES;
if (did->block) return(2);
else return(YES);
}
did = did->next;
}
return(NO);
} /* end of check_tp */
/*=====================================================================*/
void show_hom(temp, tot, blocks)
struct hom *temp;
int tot;
struct best_block *blocks;
{
int i, t, nhit, ntp, itn, itp, nblock, save_t, pearson, tp999, tn999;
int tn[NSCORE], tp[NSCORE], rank, nfrag;
int save_norm, save_strand, strand;
long locfirst, loclast;
double roc;
char save_id[25];
/* Sort by strand, seq id, normalized score (d) */
qsort(temp, tot, sizeof(struct hom), idnorm_cmp);
/* Propagate maximum normalized score to all blocks for a group */
strcpy(save_id, temp[0].seq_id); save_norm = temp[0].max_norm;
if (temp[0].frame < 0) save_strand = -1;
else save_strand = 1;
save_t = 0;
for (t=1; t<tot; t++)
{
if (temp[t].frame < 0) strand = -1; else strand = 1;
if (save_strand == strand &&
strcmp(temp[t].seq_id, save_id)==0)
{
temp[t].max_norm = save_norm;
}
else
{
strcpy(save_id, temp[t].seq_id);
save_norm = temp[t].norm;
save_t = t;
save_strand = strand;
}
}
/* Sort by max norm, strand, seq id, hom name and score */
qsort(temp, tot, sizeof(struct hom), norm_cmp);
/*--Initialize----------------------------------------------------*/
/*-----------Rank counts results that are either potential TPs or
TNs, results from the block are skipped ----------------------*/
nhit = ntp = nblock = nfrag = t = itn = itp = rank = 0;
roc = (double) 0.0;
/*----------Process first sequence -------------------------------*/
printf("\n\nTop %d sequences sorted by maximum normalized score:", MaxHit);
printf("\n(the normalized score is the Score divided by the\n");
printf("99.5 score in the block if available, or by the\n");
printf("smallest Score in the search results file for the block)");
if (SeqType == DNA)
printf("\n\nBlock Rank Frame Score Location(bp) Sequence");
else
printf("\n\nBlock Rank Frame Score Location(aa) Sequence");
if (t < tot && !temp[t].tp)
{
nhit++; tn[itn++] = rank; rank++;
if ( (LisSeq+FragSeq) > 0)
roc += (double) ntp / (double) (LisSeq + FragSeq);
if (SeqType == DNA)
{
locfirst = compute_loc(temp[t].frame, temp[t].offset+1, temp[t].seqlen);
loclast = compute_loc(temp[t].frame, temp[t].offset + temp[t].width,
temp[t].seqlen);
}
else
{
locfirst = temp[t].offset+1; loclast = temp[t].offset+temp[t].width;
}
printf("\n%d.---------------------------------------", nhit);
printf("-------------------------------------------");
printf("\n%8s %4d %2d %4d %7ld-%7ld %s %s",
temp[t].ac, temp[t].rank+1, temp[t].frame, temp[t].norm,
locfirst, loclast,
temp[t].seq_id, temp[t].title);
} /* end of not in the true positive list */
else
{ /* In the TP list */
if (temp[t].tp == 2) { nblock++; }
else { ntp++; tp[itp++] = rank; rank++; }
}
strcpy(save_id, temp[t].seq_id);
save_norm = temp[t].max_norm;
if (temp[t].frame < 0) save_strand = -1; else save_strand = 1;
save_t = t;
t++;
/*-------------------Print the rest of the hits -----------------*/
/* New hit if different seq_id or if same seq_id but different strand */
while(nhit <= MaxHit && t <= tot)
{
if (temp[t].frame < 0) strand = -1; else strand = 1;
if ( strcmp(temp[t].seq_id, save_id) != 0 ||
(strcmp(temp[t].seq_id, save_id) == 0 &&
strand != save_strand) ) /* new group */
{
if (blocks != NULL && !temp[save_t].tp) /* process previous group */
{
if (blocks != NULL) check_dat(save_t, t, temp, blocks);
else printf("\nBlocks not found in database\n");
}
strcpy(save_id,temp[t].seq_id);
if (temp[t].frame < 0) save_strand = -1; else save_strand = 1;
save_t = t;
if (t < tot && !temp[t].tp)
{
nhit++; tn[itn++] = rank; rank++;
if ( (LisSeq+FragSeq) > 0)
roc += (double) ntp / (double) (LisSeq + FragSeq);
if (nhit <= MaxHit)
{
printf("\n%d.---------------------------------------", nhit);
printf("-------------------------------------------");
}
} /* end of if not a TP */
else
{
if (temp[t].tp == 2) { nblock++; }
else { ntp++; tp[itp++] = rank; rank++; }
}
}
if (nhit <= MaxHit && t < tot && !temp[t].tp)
{
if (SeqType == DNA)
{
locfirst = compute_loc(temp[t].frame,
temp[t].offset+1, temp[t].seqlen);
loclast = compute_loc(temp[t].frame, temp[t].offset + temp[t].width,
temp[t].seqlen);
}
else
{
locfirst = temp[t].offset+1; loclast = temp[t].offset+temp[t].width;
}
printf("\n%8s %4d %2d %4d %7ld-%7ld %s %s",
temp[t].ac, temp[t].rank+1, temp[t].frame, temp[t].norm,
locfirst, loclast,
temp[t].seq_id, temp[t].title);
} /* end of if not in the true positive list */
t++;
}
/*-------------------------------------------------------------------*/
/*-- For statistics calcs, notice that the "scores" in the tn[] and
tp[] arrays are really ranks & so are sorted in reverse order
compared with matodat/blastdat/fastodat results; lower value
is "higher" score --*/
if (nhit > 0) roc /= (double) nhit;
else roc = 1.0;
tn999 = (int) ( (double) (SWISS-LisSeq-FragSeq-BlkSeq) * 0.001);
if (tn999 > nhit) tn999 = nhit - 1;
if (tn999 < 0) tn999 = 0;
tp999 = 0;
for (i=0; i < ntp; i++)
if (tp[i] <= tn[tn999]) tp999++;
pearson = LisSeq + FragSeq - ntp;
i = ntp - 1;
while (pearson >= 0 && pearson < nhit &&
i >= 0 && i < ntp &&
tn[pearson] < tp[i])
{ pearson++; i--; }
/* ntp=#tps found not in block, nblock=#tps found in block,
nhit=#found not in .lis file => tns for testing purposes */
if (Fdat != NULL)
fprintf(Fdat, "%s %d %d %d %d %d %d %d %d %d %d %d %f\n",
AC, NScore, MaxHit, LisSeq, BlkSeq, FragSeq, LisSeq+FragSeq,
nblock, ntp, nhit, tp999, pearson, roc);
} /* end of show_hom */
/*======================================================================*/
/* Sort by strand, seq id, rank */
int id_cmp(t1,t2)
struct hom *t1, *t2;
{
int strand1, strand2;
strand1 = strand2 = 1;
if (t1->frame < 0) strand1 = -1;
if (t2->frame < 0) strand2 = -1;
if (strand1 != strand2) return(strand1 - strand2);
else if (strcmp(t1->seq_id, t2->seq_id) > 0) return(1);
else if (strcmp(t1->seq_id, t2->seq_id) < 0) return(-1);
else return(t1->rank - t2->rank);
} /* end of id_cmp */
/*======================================================================*/
/* Sort by strand, seq id, normalized score (d) */
int idnorm_cmp(t1,t2)
struct hom *t1, *t2;
{
int strand1, strand2;
strand1 = strand2 = 1;
if (t1->frame < 0) strand1 = -1;
if (t2->frame < 0) strand2 = -1;
if (strand1 != strand2) return(strand1 - strand2);
else if (strcmp(t1->seq_id, t2->seq_id) > 0) return(1);
else if (strcmp(t1->seq_id, t2->seq_id) < 0) return(-1);
else return(t2->norm - t1->norm);
} /* end of idnorm_cmp */
/*======================================================================*/
/* Sort by min_rank, strand, seq id, hom name */
int rank_cmp(t1,t2)
struct hom *t1, *t2;
{
int strand1, strand2;
strand1 = strand2 = 1;
if (t1->frame < 0) strand1 = -1;
if (t2->frame < 0) strand2 = -1;
if (t1->min_rank != t2->min_rank) return(t1->min_rank - t2->min_rank);
else if (strand1 != strand2) return(strand1 - strand2);
else if (strcmp(t1->seq_id, t2->seq_id) > 0) return(1);
else if (strcmp(t1->seq_id, t2->seq_id) < 0) return(-1);
else return(strcmp(t1->ac, t2->ac));
} /* end of rank_cmp */
/*======================================================================*/
/* Sort by max normalized score (d), strand, seq id, hom name */
int norm_cmp(t1,t2)
struct hom *t1, *t2;
{
int strand1, strand2;
strand1 = strand2 = 1;
if (t1->frame < 0) strand1 = -1;
if (t2->frame < 0) strand2 = -1;
if (t1->max_norm != t2->max_norm) return(t2->max_norm - t1->max_norm);
else if (strand1 != strand2) return(strand1 - strand2);
else if (strcmp(t1->seq_id, t2->seq_id) > 0) return(1);
else if (strcmp(t1->seq_id, t2->seq_id) < 0) return(-1);
else return(strcmp(t1->ac, t2->ac));
} /* end of norm_cmp */
/*======================================================================*/
void print_blurb()
{
FILE *fstp;
char line[MAXLINE];
if ((fstp=fopen("blksrch.stp", "r")) == NULL)
{
printf("\n========================================");
printf("=======================================");
printf("\nSearch results from the BLOCKS e-mail searcher.");
printf("\nPlease report problems to jorja@sparky.fhcrc.org");
printf(", include your query\nand this output.");
printf(" To obtain help, send the word HELP on a single");
printf("\nline to blocks@howard.fhcrc.org");
printf("\n========================================");
printf("=======================================");
printf("\nCopyright (c) 1992 by the Fred Hutchinson Cancer Research Center");
printf("\nIf you use BLOCKS in your research, please cite:");
printf("\nSteven Henikoff and Jorja G. Henikoff,");
printf(" Automated assembly of protein");
printf("\nblocks for database searching, NAR 19:23 (1991), 6565-6572.");
/* printf("\n2. James C. Wallace and Steven Henikoff,");
printf(" PATMAT: a searching and extraction");
printf("\n program for sequence, pattern and block queries");
printf(" and databases,");
printf("\n CABIOS 8:3 (1992), 249-254.");
*/
printf("\n========================================");
printf("=======================================");
printf("\nEach numbered result consists of one");
printf(" or more blocks from a PROSITE group");
printf("\nfound in the query sequence. One set");
printf(" of the highest-scoring blocks that");
printf("\nare in the correct order and separated");
printf(" by distances comparable to the BLOCKS");
printf("\ndatabase is selected for analysis.");
printf(" If this set includes multiple blocks");
printf("\nthe probability that the lower scoring");
printf(" blocks support the highest scoring");
printf("\nblock is reported. Maps of the database");
printf(" blocks and query sequence are shown:");
printf("\n < indicates the sequence has been");
printf(" truncated to fit the page");
printf("\n : indicates the minimum distance");
printf(" between blocks in the database");
printf("\n . indicates the maximum distance");
printf(" between blocks in the database");
printf("\nThe maps are aligned on the highest");
printf(" scoring block. The alignment of the");
printf("\nquery sequence with the sequence");
printf(" closest to it in the BLOCKS database");
printf("\nis shown. Upper case in the query");
printf(" sequence indicates at least one");
printf("\noccurrence of the residue in that");
printf(" column of the block.");
printf("\n========================================");
printf("=======================================");
printf("\n");
}
else
{
while (!feof(fstp) && fgets(line, MAXLINE, fstp) != NULL)
printf("%s", line);
fclose(fstp);
}
} /* end of print_blurb */
/*======================================================================*/
/* Look up this group in blocks.dat */
/*======================================================================*/
void check_dat(min_t, max_t, results, fblock)
int min_t, max_t;
struct hom *results;
struct best_block *fblock;
{
int nblock;
/*---- Add the query as the last sequence in each block in the list ---*/
nblock = add_query(min_t, max_t, results, fblock);
/*--- Map the blocks if more than one in the hit ----*/
if (nblock > 1) map_blocks(min_t, max_t, results, fblock);
else printf("\n");
/*---Determine the closest block seq to the query seq & display them --*/
align_blocks(fblock);
} /* End of check_dat */
/*=====================================================================
Read the blocks database
Truncates blocks wider than MAX_WIDTH
=======================================================================*/
struct best_block *read_block(fblk, ac)
FILE *fblk;
char *ac;
{
char line[MAXLINE], *ptr, *ptr1;
struct best_block *block;
int done, aclen;
done = NO;
block = NULL;
aclen = strlen(ac);
while (!done && !feof(fblk) && fgets(line, MAXLINE, fblk) != NULL)
{
if (strncmp(line, "AC ", 5) == 0 &&
strncmp(line+5, ac, aclen) > 0)
done = YES;
else if (strncmp(line, "AC ", 5) == 0 &&
strncmp(line+5, ac, aclen) == 0)
{
block = (struct best_block *) malloc(sizeof(struct best_block));
if (block == NULL)
{ printf("\nOUT OF MEMORY\n\n"); exit(-1); }
block->query = block->anchor = NO;
block->next_block = NULL;
block->dat_pos = ftell(fblk);
strncpy(block->ac, line+5, aclen + 1); block->ac[aclen + 1] = '\0';
if (block->ac[aclen] == ';') block->ac[aclen] = '\0';
block->minprev = block->maxprev = -1;
block->query = NO; block->rank = 9999;
ptr = strtok(line+12, "(");
if (ptr != NULL)
{
ptr = strtok(NULL, ",");
if (ptr != NULL)
{
block->minprev = atoi(ptr);
ptr = strtok(NULL, ")");
if (ptr != NULL) block->maxprev = atoi(ptr);
}
}
}
else if (block != NULL && strncmp(line, "BL ", 5) == 0)
{
block->s995 = 0;
ptr=strstr(line,"99.5\%=");
if (ptr != NULL)
{
ptr1 = strtok(ptr, "="); ptr1=strtok(NULL, "\n\r");
block->s995 = atoi(ptr1);
}
block->strength = 0;
ptr=strstr(line,"strength=");
if (ptr != NULL)
{
ptr1 = strtok(ptr, "="); ptr1=strtok(NULL, "\n\r");
block->strength = atoi(ptr1);
}
fill_block(fblk, block);
return(block);
}
}
return(NULL); /* no block found! */
} /* end of read_block */
/*====================================================================
Fill up the block structure
=======================================================================*/
void fill_block(fblk, block)
FILE *fblk;
struct best_block *block;
{
int done, i, n, cluster, ncluster;
char line[MAXLINE], ctemp[MAXLINE], *ptr, *ptr1;
block->nseq = block->width = 0;
cluster = ncluster = 0;
done=NO;
while (!done && !feof(fblk) && fgets(line,MAXLINE,fblk) != NULL)
{
if (strlen(line) == 1) /* blank line => new cluster */
{
/* Set #seqs in cluster to seqs in previous cluster */
if (ncluster > 0)
for (n=0; n<block->nseq; n++)
if (block->cluster[n] == cluster) block->ncluster[n] = ncluster;
cluster++; ncluster = 0;
}
else if ((int) strlen(line) > (int) 1)
{
if (strncmp(line, "//", 2) == 0) done=YES;
else if ((int) strlen(line) > 20)
{
ptr=strtok(line, "(");
/* need to strip leading spaces off ptr here */
strcpy(ctemp, ptr);
i=0;
while (ctemp[i] == ' ') i++;
strcpy(block->name[block->nseq], ctemp+i);
ptr=strtok(NULL, ")");
block->offset[block->nseq] = atoi(ptr);
ptr1=strtok(NULL, "\n\r");
i=0;
while (ptr1[i] == ' ') i++;
ptr = strtok(ptr1+i, " \t\r\n");
/* Truncate wide blocks */
if (strlen(ptr) > MAX_WIDTH) ptr[MAX_WIDTH - 1] = '\0';
strcpy(block->aa[block->nseq], ptr);
block->cluster[block->nseq] = cluster;
ncluster++; /* # seqs in current cluster */
block->width = (int) strlen(block->aa[block->nseq]);
block->nseq++;
}
}
}
/* Compute weights for the last cluster */
if (ncluster > 0)
for (n=0; n<block->nseq; n++)
if (block->cluster[n] == cluster) block->ncluster[n] = ncluster;
} /* end of fill_block */
/*=======================================================================
Add the query sequence as the last sequence in each block for
which it has a hit.
==========================================================================*/
int add_query(min_t, max_t, results, fblock)
int min_t, max_t;
struct hom *results;
struct best_block *fblock;
{
struct best_block *block, *maxblock;
int i, t, j, imin, nblock, mblock, lastrank, one;
char ac[10], pline[MAXLINE], tline[40];
struct temp *temp;
double prob;
/*------- Need to know how many possible blocks there are ---*/
/*--------Clear out any previous sequence information -------*/
nblock=0;
block = fblock;
while (block != NULL)
{
nblock++;
block->name[block->nseq][0] = '\0';
block->offset[block->nseq] = 0;
block->cluster[block->nseq] = -1;
block->ncluster[block->nseq] = -1;
block->aa[block->nseq][0] = '\0';
if (block->query) block->nseq--;
block->query = NO;
block = block->next_block;
}
if (nblock == 1) one = YES;
else one = NO;
/*------ Sort this group of hits by rank now --------------*/
temp = (struct temp *) malloc((max_t-min_t)*sizeof(struct temp));
if (temp == NULL)
{
printf("\nOUT OF MEMORY\n\n"); exit(-1);
}
for (i=0; i<max_t-min_t; i++)
{
temp[i].value = results[i+min_t].rank;
temp[i].index = i+min_t; temp[i].flag = 0;
}
qsort(temp, max_t-min_t, sizeof(struct temp), tempcmp);
/*----------- Piece together a set of compatible hits, highest
rank first and compute probability of multiple blocks ----------*/
prob = 1.0; maxblock = NULL; pline[0] = '\0';
mblock = 0; /* number of blocks in the map */
lastrank = 0; /* rank of previous block in the map */
for (i=0; i<max_t-min_t; i++)
{
t = temp[i].index;
block = fblock;
while (block != NULL && t < max_t)
{
DBLen = results[t].seqlen;
strcpy(ac, results[t].ac);
if (strcmp(ac, block->ac) == 0 && !block->query &&
(i==0 ||
distance_okay(results[t].offset, fblock, block)) )
{
/*--- Adds the query to the block ------*/
strncpy(block->name[block->nseq], results[t].seq_id, SNAMELEN);
block->name[block->nseq][SNAMELEN] = '\0';
block->offset[block->nseq] = results[t].offset;
block->cluster[block->nseq] = -1;
block->ncluster[block->nseq] = 1;
/* Search alignment may be shorter than block ...
pad with spaces */
if ((int) strlen(results[t].aa) < block->width )
{
for (j = strlen(results[t].aa); j < block->width; j++)
results[t].aa[j] = ' ';
results[t].aa[block->width] = '\0';
}
strcpy(block->aa[block->nseq], results[t].aa);
block->query = results[t].score;
block->rank = results[t].rank;
block->nseq++;
results[t].map_flag = YES;
/*--- Compute probability -----*/
if (i==0)
{
maxblock = block;
maxblock->anchor = YES;
lastrank = temp[i].value;
}
else if (i>0 && NRank > 0.0 && DBLen > 0.0)
{
mblock++; /* number of supporting blocks so far */
block->anchor = NO;
/* blksort calcs: compute prob a block of this family could rank this high
n = (double) DBType*DBLen*NBlock-lastrank;
r = (double) temp[i].value-lastrank;
p = (double) nblock-mblock;
prob *= hypergeo(n, r, p, 1.0);
*/
/* Probability supporting block ranks this high */
prob *= (double) (temp[i].value + 1) / NRank;
/* Compute prob this block is this far from the anchor block */
/* DBLen is always in protein units, distance is always
in protein units no need to multiply by DBType to convert to
correct units. Min dist = -1; Max dist = distance() */
prob *= (double) (distance(maxblock,block)+1) / DBLen;
strcat(pline, block->ac); strcat(pline, " ");
lastrank = temp[i].value;
}
}
/* blksort calcs else if (strcmp(ac, block->ac) == 0)
nblock--; block doesn't fit
*/
block = block->next_block;
}
}
if (prob > 1.0) prob = 1.0;
if (prob < 0.0) prob = 0.0;
if (maxblock != NULL && strlen(pline))
{
printf("\n\nP<%6.2g for ", prob);
i=0;
while (i < (int) strlen(pline))
{
imin=36;
if (i+imin > (int) strlen(pline)) imin = (int) strlen(pline) - i;
strncpy(tline, pline+i, imin); tline[imin] = '\0';
if (i==0) printf("%s", tline);
else printf("\n %s", tline);
i += 36;
}
printf("in support of %s", maxblock->ac);
}
else if (maxblock != NULL)
{ printf("\n\nNo P-value computed for single block hits."); }
else
{ printf("\n\nERROR: Anchor block not found in blocks database."); }
/*----- multimat.dat statistics ------------------------------*/
for (i=0; i<max_t-min_t; i++)
{
t = temp[i].index;
if (results[t].map_flag && Fout != NULL)
fprintf(Fout, "%s %s %d %d %d %d %d %d %6.2g %s\n",
results[t].ac, results[t].seq_id, results[t].rank + 1,
results[t].frame, results[t].score, results[t].offset + 1,
nblock, mblock+1, prob, results[t].aa);
}
free(temp);
return(mblock+1);
} /* end of add_query */
/*===================================================================
Compute probability from hypergeometric distribution:
Assume there are n objects, p of one kind and n-p of another.
Then when r objects are selected at random from among the n
objects, the prob. that k of the r are of type p is:
( p ) ( n-p ) p! (n-p)!
( k ) ( r-k ) ----- * -----------------------
k!(p-k)! (r-k)!( (n-p)-(r-k) )!
----------------- = ---------------------------------
( n ) n!
( r ) ----------
r!(n-r)!
p! r! (n-p)! (n-r)!
= -------- * -------- * ---------- * -----------------
k!(p-k)! (r-k)! n! ( (n-r)-(p-k) )!
Which can be simplified to a product of these expressions:
1. k terms:
p*(p-1)*(p-2)*...(p-k+1) / k!
2. k terms:
r*(r-1)*(r-2)*...(r-k+1)
3. p terms:
1/(n*(n-1)*(n-2)*...(n-p+1))
4. p-k terms:
(n-r)*(n-r-1)*....(n-r-(p-k)+1)
=====================================================================*/
double hypergeo(n, r, p, k)
double n, r, p, k;
{
double prob;
int i;
prob = 1.0;
for (i=0; i<k; i++) prob *= (double) (p-i) * (r-i) / (i+1);
/* for (i=0; i< (p-k); i++) prob *= (double) (n-r-i);
for (i=0; i<p; i++) prob /= (double) (n-i);
*/
for (i=0; i<(p-k); i++) prob *= (double) (n-r-i)/(n-i);
for (i=(p-k); i<p; i++) prob /= (double) (n-i);
return(prob);
} /* end of hypergeo */
/*====================================================================
Check if this block will fit between existing blocks
Minimum distance between blocks is assumed to be 0 and
maximum distance is minprev+maxprev = 2xthe average distance
between the blocks in sequences in the blocks database.
=====================================================================*/
int distance_okay(offset, fblock, nblock)
int offset;
struct best_block *fblock, *nblock;
{
struct best_block *block, *lblock, *rblock;
int mindist, maxdist;
lblock = rblock = NULL;
block = fblock;
while (block != NULL && rblock == NULL)
{
if (block->query) /* query added to this block already */
{
if (strcmp(block->ac, nblock->ac) < 0)
lblock = block; /* closest block on the left */
else if (strcmp(block->ac, nblock->ac) > 0 && rblock == NULL)
rblock = block; /* closest block on the right */
}
block = block->next_block;
}
/*--------- Check the distance to the left --------------------*/
if (lblock != NULL)
{
mindist = maxdist = lblock->offset[lblock->nseq-1] + lblock->width;
block = lblock->next_block;
while (block != NULL && block != nblock)
{
if (strcmp(block->ac, nblock->ac) < 0)
{
mindist += block->width;
maxdist += prev_dist(block) + block->width;
}
block = block->next_block;
}
if (mindist-2 > offset) return(NO); /* allow overlap of 2 */
if (maxdist + prev_dist(nblock) < offset) return(NO);
}
/*----------Check the distance to the qright ---------------------*/
if (rblock != NULL)
{
mindist = maxdist = offset + nblock->width;
block = nblock->next_block;
while (block != NULL && block != rblock)
{
if (strcmp(block->ac, rblock->ac) < 0)
{
mindist += block->width;
maxdist += prev_dist(block) + block->width;
}
block = block->next_block;
}
if (mindist-2 > rblock->offset[rblock->nseq-1]) return(NO);
if (maxdist + prev_dist(rblock) < rblock->offset[rblock->nseq-1])
return(NO);
}
return(YES);
} /* end of distance_okay */
/*===================================================================
Compute maximum distance preceding a block
=====================================================================*/
int prev_dist(block)
struct best_block *block;
{
int dist;
/*
dist = block->minprev + block->maxprev;
*/
dist = 2 * block->maxprev;
if (dist < 1) dist = 1;
return(dist);
} /* end of prev_dist */
/*===================================================================
Compute maximum allowable distance between two blocks
====================================================================*/
int distance(fromblock, toblock)
struct best_block *fromblock, *toblock;
{
int dist, maxdist;
struct best_block *block, *lblock, *rblock;
if (fromblock == NULL || toblock == NULL) return(-1);
if (strcmp(fromblock->ac, toblock->ac) < 0)
{ lblock=fromblock; rblock=toblock;
maxdist = DBLen-(fromblock->offset[fromblock->nseq-1]+fromblock->width);
}
else
{ lblock=toblock; rblock=fromblock;
maxdist = fromblock->offset[fromblock->nseq-1];
}
/* For BLIMPS matrix search, DBLen is always in protein units
if (DBType*maxdist > DBLen) maxdist = (int) DBLen/DBType;
*/
if (maxdist > DBLen) maxdist = (int) DBLen;
dist = 0;
block=lblock->next_block;
while (block != NULL && block != rblock)
{
dist += prev_dist(block) + block->width;
block = block->next_block;
}
dist += prev_dist(rblock);
/* Don't want possible distance to be larger than possible
sequence positions! */
if (dist > maxdist) dist = maxdist;
return(dist);
} /* end of distance */
/*====================================================================
Map a path of blocks
The mapping is done in three phases:
1. The blocks in the database are mapped from the 1st position of
the first block to the last position of the last block. This
range divided by MAXMAP (number of columns for the map to occupy
when printed) determines the scale of the map.
2. The blocks in the query sequence that are consistent with the
database are mapped to the same scale as the database blocks.
The two maps are aligned on the highest scoring block (maxblock).
The query map may occupy more than MAXMAP columns and may have
to be truncated or compressed. A maximum of MAXLINE characters
are mapped before compression.
3. The blocks in the query sequence that are not consistent with the
database are mapped to the same scale. Since these blocks may
overlap one another, they may take more than one line. Each line
is aligned with the left end of the map in 2.
=======================================================================*/
void map_blocks(min_t, max_t, results, fblock)
int min_t, max_t;
struct hom *results;
struct best_block *fblock;
{
int totblks, maxdist, i, imin, imax, pos, maxspot, firstpos, lastpos;
int spot, qspot, firstspot, lastspot, t, done, qleft, qright;
double scale;
struct best_block *block, *maxblock;
char ctemp[20], dbline[MAXLINE], qline[MAXLINE], pline[MAXLINE];
totblks = maxdist = 0;
maxblock = NULL;
/*--- Determine the scale for the database blocks (# aas per space) ---*/
block = fblock;
while (block != NULL)
{
totblks++;
if (block != fblock)
maxdist += block->maxprev; /* distance from last block */
maxdist += block->width; /* width of this block */
if (block->query && block->anchor) maxblock = block;
block = block->next_block;
}
scale = (double) maxdist/MAXMAP; /* max of 60 spaces per line */
for (spot=0; spot<MAXLINE; spot++)
{ dbline[spot] = qline[spot] = ' '; }
/*--- 1. Now map the database blocks ------------------------------*/
spot=maxspot=0;
block = fblock;
while (block != NULL)
{
if (block != fblock)
{
imin = (int) ((double) 0.5+block->minprev/scale);
imax = (int) ((double) 0.5+(block->maxprev-block->minprev)/scale);
for (i=spot; i < spot+imin; i++) dbline[i] = ':';
spot += imin;
for (i=spot; i < spot+imax; i++) dbline[i] = '.';
spot += imax;
}
imin = (int) ((double) 0.5+block->width/scale);
if (imin < 1) imin = 1;
for (i=spot; i < spot+imin; i++) strncpy(dbline+i, block->ac+ACLen, 1);
spot += imin;
if (block==maxblock) maxspot=spot;
block = block->next_block;
}
dbline[spot] = '\0';
printf("\n |-----%5d residues----|",
(int) ((double) 0.5+(scale*25.0)) );
printf("\n%-20s ", AC); /* need SNAMELEN here to line up */
printf("%s", dbline);
/*---- 2. Now map the query sequence if there is one ----------------*/
/*---Line up the highest scoring query block with the database
pos = sequence position,
spot=scaled pos in qline , qspot=scaled pos of anchor block in query,
maxspot=scaled pos of anchor block in block map */
/* firstpos & lastpos are first & last sequence positions mapped into qline
firstspot is position of firstpos in qline, etc. */
spot=qspot=pos=0;
firstpos = lastpos = firstspot = lastspot = -1;
block = fblock;
while (block != NULL)
{
if (block->query)
{
if (firstpos < 0) firstpos = block->offset[block->nseq-1];
lastpos = block->offset[block->nseq-1] + block->width;
imin = (int) ((double) 0.5+(block->offset[block->nseq-1]-pos)/scale);
if (imin >= 0) /* skip over blocks in wrong order */
{
if (spot+imin > MAXLINE && spot+1 < MAXLINE)
qline[spot++] = '<';
else
{
for (i=spot; i<spot+imin; i++) qline[i] = ':';
spot += imin;
}
imin = (int) ((double) 0.5+block->width/scale);
if (imin < 1) imin = 1;
if (spot+imin < MAXLINE)
{
for (i=spot; i<spot+imin; i++)
strncpy(qline+i, block->ac+ACLen, 1);
spot += imin;
}
pos = block->offset[block->nseq-1] + block->width;
if (block == maxblock) qspot = spot;
}
}
block = block->next_block;
}
qline[spot] = '\0';
/* Line up qline/qspot with dbline/maxspot */
strncpy(ctemp, results[min_t].seq_id, SNAMELEN); ctemp[SNAMELEN] = '\0';
printf("\n%-20s ", ctemp);
if (qspot <= maxspot)
{
qleft = 0; firstspot = maxspot - qspot;
for (i=qspot; i<maxspot; i++) printf(" ");
}
else
{
qleft = qspot - maxspot; firstspot = 0;
qline[0] = '<';
}
if (((int) strlen(qline) - qspot) <= ((int) strlen(dbline) - maxspot))
qright = (int) strlen(qline);
else
qright = qspot + (int) strlen(dbline) - maxspot;
strncpy(pline, qline+qleft, qright-qleft+1); pline[qright-qleft+1] = '\0';
if (qleft > 0 && (pline[0] == '.' || pline[0] == ':')) pline[0] = '<';
if (qright != (int) strlen(qline)) pline[(int) strlen(pline)-1] = '>';
printf("%s\n", pline);
/*--------3. Now map the inconsistent hom hits wrt consistent ones--*/
/* Take the hits as they come, as many as will fit on a line */
done = NO;
while (!done)
{
done=YES; spot=0;
for (i=0; i<MAXLINE; i++) qline[i] = ' ';
lastspot = 0;
for (t=min_t; t<max_t; t++)
{
if (!results[t].map_flag)
{
imin = (int) ((double) 0.5+(results[t].offset)/scale);
if (results[t].offset < firstpos) spot = firstspot - imin;
else spot = firstspot + imin;
if (spot >= 0 && spot < MAXLINE && qline[spot] == ' ')
{
imin = (int) ((double) 0.5+results[t].width/scale);
if (imin < 1) imin = 1;
i = spot;
while (i < spot+imin && i < MAXLINE)
{
strncpy(qline+i, results[t].ac+ACLen, 1);
if (i > lastspot) lastspot = i;
i++;
}
done = NO;
results[t].map_flag = YES;
}
}
} /* end of a line of output */
qline[lastspot+1] = '\0';
if (!done && strlen(qline))
{
printf("%-20s %s\n", ctemp, qline); /* ctemp is hit name */
}
} /* end of inconsistent hits */
printf("\n");
} /* end of map_blocks */
/*========================================================================
Align the query sequence with the sequence closest to it in
each block for which it has a hit.
datline from blocks database
homline from blimps search results
=========================================================================*/
void align_blocks(fblock)
struct best_block *fblock;
{
struct best_block *block;
int s, i, itemp, spot, bspot, datmin, datmax, homdist;
char datline[MAXLINE], homline[MAXLINE], blkline[MAXLINE];
char barline[MAXLINE], saveac[MAXAC+1];
char ctemp[MAXLINE];
for (i=0; i<MAXLINE; i++)
{ datline[i] = homline[i] = blkline[i] = barline[i] = ' ';}
bspot=spot=datmin=datmax=homdist=0;
saveac[0] = '\0';
for (i=0; i<ACLen+1; i++) strcat(saveac, " ");
block=fblock;
while (block != NULL)
{
if (block->query)
{
datmin += block->minprev;
datmax += block->maxprev;
homdist = block->offset[block->nseq-1] - homdist;
consensus(block);
/*>>> if aborts on call to closest_seq(), reduce MAXSEQS in motifj.h
or set s = 0
<<<*/
s = closest_seq(block);
bspot = block->width;
/*--- 28 spaces for name & offset on datline & homline;
27 spaces for distance info on blkline ----*/
if (bspot < 27) bspot = 27; /* min blkline width */
if ((spot + bspot + 28) > 80)
{
itemp = spot;
if (bspot > itemp) itemp = bspot;
homline[itemp]= datline[itemp]= blkline[itemp]= barline[itemp]= '\0';
printf("\n%s", blkline);
printf("\n%s", datline); printf("\n%s", barline);
printf("\n%s\n", homline);
for (i=0; i<MAXLINE; i++)
{ datline[i] = homline[i] = blkline[i] = barline[i] = ' '; }
spot=0;
}
sprintf(ctemp, "%-20s", block->ac); /* SNAMELEN */
strncpy(blkline+spot, ctemp, strlen(ctemp));
sprintf(ctemp, "%-20s", block->name[s]); /* SNAMELEN */
strncpy(datline+spot, ctemp, strlen(ctemp));
sprintf(ctemp, "%-20s", block->name[block->nseq-1]);
strncpy(homline+spot, ctemp, strlen(ctemp));
spot+=SNAMELEN + 1; /* name width + 1 */
/*
strncpy(blkline+spot, block->ac, strlen(block->ac));
strncpy(datline+spot, block->name[s], strlen(block->name[s]));
strncpy(homline+spot, block->name[block->nseq-1],
strlen(block->name[block->nseq-1]));
*/
sprintf(ctemp, "%c<->%c", saveac[ACLen], block->ac[ACLen]);
strncpy(blkline+spot, ctemp, strlen(ctemp));
sprintf(ctemp, "%6d", block->offset[s]);
strncpy(datline+spot, ctemp, strlen(ctemp));
sprintf(ctemp, "%6d", block->offset[block->nseq-1] + 1);
strncpy(homline+spot, ctemp, strlen(ctemp));
spot+=9; /* offset width + 2 */
/*
blkline[spot] = saveac[ACLen];
strncpy(blkline+spot+1, "<->", 3);
blkline[spot+4] = block->ac[ACLen];
kr_itoa(block->offset[s], ctemp, 10);
strncpy(datline+spot, ctemp, strlen(ctemp));
kr_itoa(block->offset[block->nseq-1]+1, ctemp,10);
strncpy(homline+spot, ctemp, strlen(ctemp));
*/
bspot = spot;
sprintf(ctemp, "(%d,%d):%d", datmin, datmax, homdist);
/*
strncpy(blkline+bspot, "(", 1); bspot++;
kr_itoa(datmin, ctemp, 10);
strncpy(blkline+bspot, ctemp, strlen(ctemp));
bspot += strlen(ctemp);
strncpy(blkline+bspot, ",", 1); bspot++;
kr_itoa(datmax, ctemp, 10);
strncpy(blkline+bspot, ctemp, strlen(ctemp));
bspot += strlen(ctemp);
strncpy(blkline+bspot, ")", 1); bspot++;
strncpy(blkline+bspot, ":", 1); bspot++;
kr_itoa(homdist, ctemp, 10);
*/
strncpy(blkline+bspot, ctemp, strlen(ctemp));
bspot += strlen(ctemp);
for (i=0; i<block->width; i++)
{
strncpy(datline+spot, block->aa[s]+i, 1);
strncpy(homline+spot, block->aa[block->nseq-1]+i, 1);
if (strncmp(datline+spot, homline+spot, 1) == 0)
barline[spot] = '|';
spot++;
}
spot+=3; /* space between blocks on same line */
datmin = datmax = 0;
homdist = block->offset[block->nseq-1] + block->width;
strcpy(saveac, block->ac);
}
else
{ /* block not in query sequence */
datmin += block->minprev + block->width;
datmax += block->maxprev + block->width;
}
block = block->next_block;
}
itemp = spot;
if (bspot > itemp) itemp = bspot;
homline[itemp] = datline[itemp] = blkline[itemp] = barline[itemp] = '\0';
printf("\n%s", blkline);
printf("\n%s", datline); printf("\n%s", barline);
printf("\n%s\n", homline);
} /* end of align_blocks */
/*======================================================================*/
/*--- Just find the sequence closest to the last sequence in the block
based on the number of identities */
/*======================================================================*/
int closest_seq(block)
struct best_block *block;
{
int npair, s1, s2, l1, l2, i, i1, i2;
int maxscore, maxs1;
struct pair pairs[MAXSEQS*(MAXSEQS-1)/2];
npair = block->nseq;
/* Compute scores for all possible pairs of sequences */
for (s1=0; s1<block->nseq-1; s1++) /* col = 0, n-2 */
{
l1 = 0;
s2 = block->nseq-1; /* last seq in block == the query sequence */
l2 = 0;
pairs[s1].score = 0;
pairs[s1].cluster = -1;
for (i=0; i<=block->width; i++)
{
i1 = l1+i; i2 = l2+i;
/* s1 is the block seq & is all caps, but s2 is the query
seq & it may have lower case chars in it */
if (i1 >= 0 && i1 < block->width &&
i2 >= 0 && i2 < block->width &&
strncasecmp(block->aa[s1]+i1, block->aa[s2]+i2, 1) == 0)
pairs[s1].score += 1;
}
} /* end of s1 */
maxs1 = -1; maxscore = -1;
for (s1=0; s1 < block->nseq-1; s1++)
if (pairs[s1].score > maxscore)
{
maxscore = pairs[s1].score;
maxs1 = s1;
}
return(maxs1);
} /* end of cluster_seqs */
/*====================================================================*/
int tempcmp(t1, t2)
struct temp *t1, *t2;
{
return(t1->value - t2->value);
} /* end of tempcmp */
/*====================================================================*/
struct best_block *get_blocks(fblk, ac)
FILE *fblk;
char *ac;
{
struct best_block *block, *fblock, *lblock;
int nblk;
nblk = 0;
block = fblock = lblock = NULL;
/*----- Make a list of blocks in the database -----*/
while ( (block = read_block(fblk, ac)) != NULL)
{
if (nblk > 0) lblock->next_block = block;
else fblock = block;
nblk++;
lblock=block;
}
if (fblock == NULL)
printf("\nNo blocks for family %s found in database.", ac);
return(fblock);
} /* end of get_blocks */
/*======================================================================
If the search was of a matrix database, the sequence segment will be
all lower case. Change any residue that matches any sequence in the
block to upper case. The block segments are all upper case.
========================================================================*/
void consensus(block)
struct best_block *block;
{
int s, s1, i;
char seqaa[2];
s = block->nseq - 1; /* This is the query sequence */
for (s1=0; s1 < s; s1++)
for (i=0; i < block->width; i++)
{
if (strncasecmp(block->aa[s]+i, block->aa[s1]+i, 1) == 0)
{
strncpy(seqaa, block->aa[s]+i, 1);
if (seqaa[0] >= 97 && seqaa[0] <= 122)
{
seqaa[0] -= 32;
strncpy(block->aa[s]+i, seqaa, 1);
}
}
}
} /* end of consensus */
/*=========================================================================
Blimps reports seqlen in AAs for Block vs DNA search
frame 1:1 => bp 1, 2:1 => 2, 3:1 => 3
bp = frame + 3 * offset
frame -1:1 => bp QLen, -2:1 => QLen - 1, -3:l => QLen - 2
bp = frame + (QLen + 1) - 3 * offset
=========================================================================*/
int compute_loc(frame, offset, seqlen)
int frame;
int offset;
double seqlen;
{
int loc;
loc = (offset - 1) * SeqType;
if (frame < 0) loc = (seqlen * SeqType + 1) - loc;
loc += frame;
if (frame == 0) loc++; /* Blimps frame = 0 for protein */
return(loc);
} /* end of compute_loc */
/*========================================================================
If the blocks have 99.5 score in them, renormalize with it
=========================================================================*/
void re_normalize(temp, totscores, blocks)
struct hom *temp;
int totscores;
struct best_block *blocks;
{
int i;
struct best_block *block;
for (i=0; i < totscores; i++)
{
block = blocks;
while (block != NULL)
{
if (strcmp(temp[i].ac, block->ac) == 0 &&
block->s995 > 0)
{
temp[i].norm = round((float) 1000. * temp[i].score / block->s995);
temp[i].max_norm = temp[i].norm;
}
block = block->next_block;
}
}
} /* end of re_normalize */
|