diff -uNr meme_4.5.0/configure meme_4.5.0.patch_1/configure --- meme_4.5.0/configure 2010-10-14 08:47:03.000000000 +1000 +++ meme_4.5.0.patch_1/configure 2010-12-03 11:08:37.000000000 +1000 @@ -12157,16 +12157,17 @@ # Check for compiler-specific settings +LARGE_FILE_SUPPORT=`getconf LFS_CFLAGS` if test "${CC}" = "gcc"; then ac_cc_opt="-O3" ac_cc_warn="-Wall -Wno-unused" ac_cc_debug="-ggdb" - CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__" + CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__ $LARGE_FILE_SUPPORT" else ac_cc_opt="-O" ac_cc_warn="" ac_cc_debug="-g" - CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__" + CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__ $LARGE_FILE_SUPPORT" fi # Set up default libxml2 settings diff -uNr meme_4.5.0/configure.ac meme_4.5.0.patch_1/configure.ac --- meme_4.5.0/configure.ac 2010-10-07 10:01:46.000000000 +1000 +++ meme_4.5.0.patch_1/configure.ac 2010-12-03 09:54:53.000000000 +1000 @@ -60,16 +60,17 @@ AM_CONDITIONAL(WANT_STRLCPY, test ["$have_strlcpy" = no]) # Check for compiler-specific settings +LARGE_FILE_SUPPORT=`getconf LFS_CFLAGS` if test "${CC}" = "gcc"; then ac_cc_opt="-O3" ac_cc_warn="-Wall -Wno-unused" ac_cc_debug="-ggdb" - CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__" + CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__ $LARGE_FILE_SUPPORT" else ac_cc_opt="-O" ac_cc_warn="" ac_cc_debug="-g" - CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__" + CFLAGS="$ac_cc_warn -DUNIX -D__USE_FIXED_PROTOTYPES__ $LARGE_FILE_SUPPORT" fi # Set up default libxml2 settings diff -uNr meme_4.5.0/etc/tomtom-to-html.xsl meme_4.5.0.patch_1/etc/tomtom-to-html.xsl --- meme_4.5.0/etc/tomtom-to-html.xsl 2010-09-06 11:18:05.000000000 +1000 +++ meme_4.5.0.patch_1/etc/tomtom-to-html.xsl 2010-12-03 09:54:53.000000000 +1000 @@ -391,6 +391,9 @@ + + () + , @@ -571,7 +574,8 @@ - logo_

_ + align_

_ + _ + -

.png diff -uNr meme_4.5.0/scripts/meme2meme.pl meme_4.5.0.patch_1/scripts/meme2meme.pl --- meme_4.5.0/scripts/meme2meme.pl 2010-09-03 15:14:40.000000000 +1000 +++ meme_4.5.0.patch_1/scripts/meme2meme.pl 2010-12-03 10:01:12.000000000 +1000 @@ -516,11 +516,14 @@ my @motif_list = (); my $motif = {}; my $index = 0; + my $new_pssm_ref; + my $new_pspm_ref; while ($line) { #look for the start of a motif if ($line =~ m/^MOTIF\s+(.*)$/) { my @words = split(/\s+/, $1); + die("Motif $index in $file missing motif name.\n") unless scalar(@words) > 0 && $words[0] ne ""; # if there was a previous motif then add it to the list push(@motif_list, $motif) if $motif->{id}; @@ -530,22 +533,25 @@ $pssm{$letter} = []; $pspm{$letter} = []; } + $new_pssm_ref = \%pssm; + $new_pspm_ref = \%pspm; my $id = $words[0]; - my $alt = ($words[1] ne "width" ? $words[1] : ""); + my $alt = (scalar(@words) > 1 && $words[1] ne "width" ? $words[1] : ""); my $url = $url_pattern; $url =~ s/MOTIF_NAME/$id/g; #use the original id in the url $motif = {id => $id, alt => $alt, bg => \%bg, - strands => $strands, url => $url, width => 0, sites => 0, pseudo => 0, evalue => 0, - pssm => \%pssm, pspm => \%pspm}; - + strands => $strands, url => $url, width => 0, sites => 0, pseudo => 0, evalue => 0}; + + $index += 1; } #look for the start of a log odds matrix elsif ($line =~ m/log-odds matrix:\s+(.*)$/) { $line = $1; $line =~ s/\s+$//; #trim right + $motif->{pssm} = $new_pssm_ref; my %motif_params = split(/\s+/, $line); #check that we have the parameters we require die("Motif $index in $file is missing required parameter(s) in log-odds matrix.\n") @@ -569,6 +575,7 @@ elsif ($line =~ m/^letter-probability matrix:\s+(.*)$/) { $line = $1; $line =~ s/\s+$//; #trim right + $motif->{pspm} = $new_pspm_ref; my %motif_params = split(/\s+/, $line); #check that we have the parameters we require die("Motif $index in $file is missing required parameter(s) in letter-probability matrix.\n") diff -uNr meme_4.5.0/scripts/runcheck_p.in meme_4.5.0.patch_1/scripts/runcheck_p.in --- meme_4.5.0/scripts/runcheck_p.in 2010-09-03 15:11:51.000000000 +1000 +++ meme_4.5.0.patch_1/scripts/runcheck_p.in 2010-12-03 10:04:14.000000000 +1000 @@ -1,40 +1,8 @@ #!/bin/sh -# runcheck - for testing of meme and mast +# runcheck_p - for testing of the parallel version of MEME. # # $Id: runcheck 1537 2007-01-12 21:48:56Z cegrant $ -# -# $Log$ -# Revision 1.7 2006/03/08 20:50:11 nadya -# merge chamges from v3_5_2 branch -# -# Revision 1.6.4.2 2006/01/24 02:24:29 nadya -# EXT is obsoleted by using printf() -# -# Revision 1.6.4.1 2006/01/24 02:04:55 nadya -# source meme_config to set needed environment -# make output table-like -# -# Revision 1.6 2005/10/07 02:54:28 nadya -# clean debugging output, fix typo -# -# Revision 1.5 2005/10/04 20:03:23 nadya -# changes to accommodate sh on solaris: -# substitute array with awk parsing and initialize err. -# -# Revision 1.4 2005/10/04 00:02:04 nadya -# change to "-f", the option "-e" does not run solaris -# -# Revision 1.3 2005/10/02 05:09:39 nadya -# add extension ".bin" to executables names for meme and mast -# -# Revision 1.2 2005/09/01 00:42:05 nadya -# call executables from $topdir befiore they are installed -# update sed regexpr -# -# Revision 1.1 2005/08/31 04:51:55 nadya -# create script and target for "make check" -# # set meme environment top=`pwd`/.. @@ -44,7 +12,6 @@ testdir=$top/tests fi meme=$top/src/parallel/meme_p -mast=$top/src/mast echo "****************************************" echo "Now Running Test for Parallel Version..." @@ -57,35 +24,9 @@ exit 1 fi -if [ ! -x $mast ] ; then - echo "$mast does not exist or is not executable" - echo "Run 'make' before running 'make check'" - exit 1 -fi - #################################################################### -# define functions +# define functions #################################################################### -# run mast tests -func_run_mast () { - echo "" - echo " Mast test for " - echo " ---------------------------------" - echo " Dataset Model result " - echo " ---------------------------------" - - for mod in $models - do - memefile=meme/meme.$dset.$mod$suffix - mastfile=mast/mast.$dset.$mod$suffix - database="$dir/common/$dset.s" - params="$dir/$memefile $database -oc $dirout/mast -nohtml -nostatus" - $mast $params - mv $dirout/mast/mast.xml $dirout/$mastfile - func_diff $mastfile "mast" - done -} - # run meme tests func_run_meme () { meme="@mpirun@ 2 $meme " @@ -117,22 +58,8 @@ e10="/Last updated/d" e11="/MPI/d" e12="/^$/d" - e13="/^mast /d" sed -e "$e1" -e "$e2" -e "$e3" -e "$e4" -e "$e5" -e "$e6" -e "$e7" \ - -e "$e8" -e "$e9" -e "$e10" -e "$e11" -e "$e12" -e "$e13" $1 > $1.sed -} - -func_xml_sed() { - e1='s///' - e2='s|[^<]*||' - e3='s|[^<]*||' - e4='s|[^<]*||' - e5='s|||' - e6='s/last_mod_date="[^"]*"/last_mod_date=""/' - e7='s/source="[^"]*"/source=""/' - e8='s/ $1.sed + -e "$e8" -e "$e9" -e "$e10" -e "$e11" -e "$e12" $1 > $1.sed } # checks two files for differences. Arguments - a filename @@ -154,13 +81,8 @@ echo "$str" "SKIPPED" return fi - if [ "$2" = "mast" ]; then - func_xml_sed $f1 - func_xml_sed $f2 - else - func_sed $f1 - func_sed $f2 - fi + func_sed $f1 + func_sed $f2 num=`diff $f1.sed $f2.sed | wc -l` status=$? if [ $status -eq 0 -a $num -eq 0 ]; then @@ -179,13 +101,6 @@ params="-text -dna -revcomp -nostatus -nmotifs 2 -minw 8" dir=$top/tests dirout=$top/tests/results - if [ ! -d $dirout/mast ]; then - cmd=`mkdir -p $dirout/mast` - status=$? - if [ $status -eq 1 ] ; then - exit 1 - fi - fi if [ ! -d $dirout/meme ]; then cmd=`mkdir -p $dirout/meme` status=$? @@ -225,7 +140,6 @@ # run tests func_set func_run_meme -func_run_mast func_clean exit $err diff -uNr meme_4.5.0/src/background.c meme_4.5.0.patch_1/src/background.c --- meme_4.5.0/src/background.c 2010-09-03 15:14:05.000000000 +1000 +++ meme_4.5.0.patch_1/src/background.c 2010-12-03 09:34:19.000000000 +1000 @@ -494,7 +494,7 @@ // get the total probability of this prefix summed over all suffixes if (wa % index_alen == 0) { int j; - for (j=0, total_w_prob=0; jmaxima; int n_psites = model->nsites_dis; fprintf(stdout, - "PREDICTED SITES AFTER EM FROM STARTING POINT WITH W = %i AND" + "\nPREDICTED SITES AFTER EM FROM STARTING POINT WITH W = %i AND" " NSITES = %f:\n", s_point->w0, s_point->nsites0); //model->w, n_psites); print_site_array(pred_sites, n_psites, stdout, model->w, dataset); double sig = dataset->objfun==Pv ? model->logpv : model->logev; @@ -617,6 +617,7 @@ /* initialize the new motif */ strcpy(model->cons0, s_point->cons0); + if (VERBOSE) { fprintf(stderr, "\nStarting point: %s\n", model->cons0); } w0 = model->w = model->pw = s_point->w0; init_theta(model->theta, s_point->e_cons0, w0, dataset->map,dataset->alength); diff -uNr meme_4.5.0/src/pmp_bf.c meme_4.5.0.patch_1/src/pmp_bf.c --- meme_4.5.0/src/pmp_bf.c 2010-09-03 15:11:50.000000000 +1000 +++ meme_4.5.0.patch_1/src/pmp_bf.c 2010-12-03 09:54:08.000000000 +1000 @@ -100,7 +100,8 @@ strcat(usage, " --pseudocount (default=0.1)\n"); strcat(usage, " --ustar \n"); // TLB; create uniform star tree strcat(usage, " --verbosity [1|2|3|4] (default 2)\n"); - strcat(usage, "\n"); + strcat(usage, "\n Prints the FP and FN rate at each of 10000 score values.\n"); + strcat(usage, "\n Output format: [ score FPR TPR ]+\n"); // Parse the command line. if (simple_setopt(argc, argv, n_options, pmp_options) != NO_ERROR) { diff -uNr meme_4.5.0/src/tomtom.c meme_4.5.0.patch_1/src/tomtom.c --- meme_4.5.0/src/tomtom.c 2010-09-03 15:14:07.000000000 +1000 +++ meme_4.5.0.patch_1/src/tomtom.c 2010-12-03 09:54:08.000000000 +1000 @@ -74,7 +74,8 @@ " the motifs are not listed in any particular order -->\n" "\n" "\n" -"\n" +"\n" "\n" @@ -123,6 +124,7 @@ * An object for storing information about a file containing a set of motifs **************************************************************************/ struct motif_db { + int id; //an id assigned to the database based on the order it was listed char *source; char *name; time_t last_mod; @@ -193,7 +195,7 @@ * Makes a copy of the strings so there is no confusion on what can be * freed later. **************************************************************************/ -static MOTIF_DB_T* create_motif_db(char *source, char *name, time_t now) { +static MOTIF_DB_T* create_motif_db(int id, char *source, char *name, time_t now) { MOTIF_DB_T *db; assert(source != NULL); @@ -201,6 +203,7 @@ db = mm_malloc(sizeof(MOTIF_DB_T)); memset(db, 0, sizeof(MOTIF_DB_T)); + db->id = id; copy_string(&(db->source), source); copy_string(&(db->name), name); //if the input is from stdin, or we can't do any better than use now as the timestamp @@ -246,6 +249,7 @@ * **************************************************************************/ static char *create_logo( + MOTIF_DB_T* target_db, // Target Database MOTIF_T* target, // Target motif MOTIF_T* query, // Query motif int offset, // Offset of target rel. to query @@ -253,18 +257,31 @@ char *output_dirname // Output directory ) { char *logo_filename, *logo_path; + int num_len, num; double logo_height = LOGOHEIGHT; double logo_width = offset <= 0 ? MAX(query->length, (target->length - offset)) : MAX((query->length + offset), target->length); + // calculate the length of the id even though technically it should be 1 in a large number of cases + num_len = 1; + num = target_db->id; + if (num < 0) { + num_len++; + num = -num; + } + while (num >= 10) { + num_len++; + num /= 10; + } + // create name of logo file logo_filename = (char*) malloc( - sizeof(char) * (5 + strlen(query->id) + 1 + strlen(target->id) + 1) + sizeof(char) * (6 + strlen(query->id) + 1 + num_len + 1 + strlen(target->id) + 1) ); - sprintf(logo_filename, "logo_%s_%s", query->id, target->id); + sprintf(logo_filename, "align_%s_%d_%s", query->id, target_db->id, target->id); // create output file path logo_path = make_path_to_file(output_dirname, logo_filename); @@ -1537,10 +1554,12 @@ *****************************************************************************/ static void print_xml_query_match(FILE *file, TOMTOM_MATCH_T *match, char **buffer, int *buffer_len) { char *target_id; + int db_id; target_id = get_motif_id(match->target)+1; - fprintf(file, "\t\t\t\t\n", - replace_xml_chars2(target_id, buffer, buffer_len, 0, TRUE), (match->orientation == '+' ? "forward" : "reverse"), + db_id = match->db->id; + fprintf(file, "\t\t\t\t\n", + db_id, replace_xml_chars2(target_id, buffer, buffer_len, 0, TRUE), (match->orientation == '+' ? "forward" : "reverse"), match->offset, match->pvalue, match->evalue, match->qvalue); } @@ -1619,8 +1638,27 @@ for (i = 0; i < arraylst_size(target_dbs); ++i) { MOTIF_DB_T *db; RBNODE_T *node; + int db_id_len, prefix_len, num; + char *prefix; + db = arraylst_get(i, target_dbs); - fprintf(xml_output, "\t\tid; + if (num < 0) { + num = -num; + db_id_len++; + } + while (num >= 10) { + db_id_len++; + num /= 10; + } + + prefix_len = 2 + db_id_len + 2; + prefix = (char*)mm_malloc(sizeof(char) * prefix_len); + snprintf(prefix, prefix_len, "t_%d_", db->id); + + fprintf(xml_output, "\t\tid); fprintf(xml_output, "source=\"%s\" ", replace_xml_chars2(db->source, buffer, buffer_len, 0, TRUE)); fprintf(xml_output, "name=\"%s\" ", replace_xml_chars2(db->name, buffer, buffer_len, 0, TRUE)); fprintf(xml_output, "loaded=\"%d\" ", db->loaded); @@ -1629,7 +1667,7 @@ for (node = rbtree_first(db->matched_motifs); node != NULL; node = rbtree_next(node)) { MOTIF_T *target; target = (MOTIF_T*)rbtree_value(node); - print_xml_motif(xml_output, "t_", get_motif_id(target)+1, target, "\t\t\t", "\t", buffer, buffer_len); + print_xml_motif(xml_output, prefix, get_motif_id(target)+1, target, "\t\t\t", "\t", buffer, buffer_len); } fprintf(xml_output, "\t\t\n"); } @@ -1916,6 +1954,7 @@ // allow one of them to use stdin { BOOLEAN_T stdin_used = FALSE; + int target_db_id = 1; query_filename = argv[option_index++]; query_last_mod = now; if (strcmp(query_filename, "-") == 0) { @@ -1953,7 +1992,7 @@ die("Target file \"%s\" does not exist.\n", target_file); target_name = NULL;//make the complier happy } - db = create_motif_db(target_file, target_name, now); + db = create_motif_db(target_db_id++, target_file, target_name, now); arraylst_add(db, target_dbs); } } @@ -2012,7 +2051,7 @@ int num, i; //note num is protected location index plus 1 num = *((int*)rbtree_key(node)); - if (num < 0) continue;//check for bad numbers (note zero is accepted despite the first num being 1 + if (num < 0) break;//check for bad numbers (note zero is accepted despite the first num being 1 //because we want to remove the range down to index zero) for (i = last_index - 1; i >= num; --i) { MOTIF_T *motif; @@ -2037,17 +2076,19 @@ int i, j; ARRAYLST_T *target_motifs = arraylst_create(); { - RBTREE_T *ids; BOOLEAN_T created, printed_warning; MOTIF_T *motif; ARRAY_T *background; BOOLEAN_T has_reverse_strand; - ids = rbtree_create(compare_strings, NULL, NULL, NULL, NULL); //for each target database for (i = 0, j = 0; i < arraylst_size(target_dbs); ++i) { + RBTREE_T *ids; MOTIF_DB_T *db; + //for duplicate checking within the database (other databases may share the same identifiers) + ids = rbtree_create(compare_strings, NULL, NULL, NULL, NULL); db = (MOTIF_DB_T*)arraylst_get(i, target_dbs); + //record the starting index of the motifs for this database db->list_index = j; background = NULL; has_reverse_strand = FALSE; @@ -2064,6 +2105,7 @@ &background ); + //record the number of motifs in this database db->loaded = arraylst_size(target_motifs) - db->list_index; //check that the background is the same @@ -2097,7 +2139,7 @@ bg_freqs_target = background; has_reverse_strand_target = has_reverse_strand; } - //check for duplicate identifiers + //check for duplicate identifiers in the same file for (; j < arraylst_size(target_motifs); ++j) { motif = (MOTIF_T*)arraylst_get(j, target_motifs); rbtree_lookup(ids, get_motif_id(motif), TRUE, &created); @@ -2120,8 +2162,9 @@ if (db->excluded) fprintf(stderr, "Warning, %d target motif(s) with duplicate ids in \"%s\" were removed.\n", db->excluded, db->name); } db->list_entries = j - db->list_index; + //destroy the set of ids for this database + rbtree_destroy(ids); }//for each target db - rbtree_destroy(ids); }//end loading motifs /********************************************** @@ -2488,7 +2531,7 @@ if (!text_only) { // create alignment logo char *logo_filename - = create_logo(target_motif, query_motif, my_match->offset, ssc, output_dirname); + = create_logo(target_db, target_motif, query_motif, my_match->offset, ssc, output_dirname); print_xml_query_match(match_output, my_match, &xml_convert_buffer, &xml_convert_buffer_len); diff -uNr meme_4.5.0/website/cgi-bin/tomtom_request.pl meme_4.5.0.patch_1/website/cgi-bin/tomtom_request.pl --- meme_4.5.0/website/cgi-bin/tomtom_request.pl 2010-09-03 15:15:20.000000000 +1000 +++ meme_4.5.0.patch_1/website/cgi-bin/tomtom_request.pl 2010-12-03 09:54:53.000000000 +1000 @@ -136,8 +136,8 @@ # write the motifs to a temporary file that will be deleted when perl exits my ($fh, $tmpname); ($fh, $tmpname) = tempfile(UNLINK => 1); - $fh->print($motif); - $fh->close(); + print $fh $motif; + close $fh; return $tmpname; }