VSEARCH 1.0.7: Fixed bug in output from chimera detection with --uchi…

…meout option, reduced default for --maxseqlength option, and removed some compiler warnings
torognes · Dec 19, 2014 · e390b73 · e390b73
1 parent b8d4828
commit e390b73
Show file tree

Hide file tree

Showing 14 changed files with 46 additions and 46 deletions.
diff --git a/README.md b/README.md
@@ -30,22 +30,22 @@ If you can't find an answer in the VSEARCH documentation, please visit the [VSEA
 
 In the example below, VSEARCH will identify sequences in the file database.fsa that are at least 90% identical on the plus strand to the query sequences in the file queries.fsa and write the results to the file alnout.txt.
 
-`./vsearch-1.0.6-linux-x86_64 --usearch_global queries.fsa --db database.fsa --id 0.9 --alnout alnout.txt`
+`./vsearch-1.0.7-linux-x86_64 --usearch_global queries.fsa --db database.fsa --id 0.9 --alnout alnout.txt`
 
 ## Download and install
 
 The latest releases of VSEARCH are available [here](https://github.com/torognes/vsearch/releases).
 
-Binary executables of VSEARCH are available in the `bin` folder for [GNU/Linux on x86-64 systems](https://github.com/torognes/vsearch/blob/master/bin/vsearch-1.0.6-linux-x86_64) and [Apple Mac OS X on x86-64 systems](https://github.com/torognes/vsearch/blob/master/bin/vsearch-1.0.6-osx-x86_64). These executables include support for  input files compressed by zlib and bzip2 (with files usually ending in .gz or .bz2).
+Binary executables of VSEARCH are available in the `bin` folder for [GNU/Linux on x86-64 systems](https://github.com/torognes/vsearch/blob/master/bin/vsearch-1.0.7-linux-x86_64) and [Apple Mac OS X on x86-64 systems](https://github.com/torognes/vsearch/blob/master/bin/vsearch-1.0.7-osx-x86_64). These executables include support for  input files compressed by zlib and bzip2 (with files usually ending in .gz or .bz2).
 
 Download the appropriate executable and make a symbolic link in a folder included in your `$PATH` from `vsearch` to the appropriate binary. You may use the following commands (assuming `~/bin` is in your `$PATH`):
 
 ```sh
 cd ~
 mkdir -p bin
 cd bin
-wget https://github.com/torognes/vsearch/releases/download/v1.0.6/vsearch-1.0.6-linux-x86_64
-ln -s vsearch-1.0.6-linux-x86_64 vsearch
+wget https://github.com/torognes/vsearch/releases/download/v1.0.7/vsearch-1.0.7-linux-x86_64
+ln -s vsearch-1.0.7-linux-x86_64 vsearch
 ```
 
 Substitute `linux` with `osx` in those lines if you're on a Mac.

diff --git a/doc/vsearch.1 b/doc/vsearch.1
@@ -1,5 +1,5 @@
 .\" ============================================================================
-.TH vsearch 1 "December 14, 2014" "version 1.0.6" "USER COMMANDS"
+.TH vsearch 1 "December 19, 2014" "version 1.0.7" "USER COMMANDS"
 .\" ============================================================================
 .SH NAME
 vsearch \(em chimera detection, clustering, dereplication, masking, pairwise alignment, searching, shuffling and sorting of amplicons from metagenomic projects.
@@ -1411,6 +1411,9 @@ Fixes a minor bug with --allpairs_global and --acceptall options.
 .TP
 .BR v1.0.6\~ "released December 14th, 2014"
 Fixes a memory allocation bug in chimera detection (--uchime_ref option).
+.TP
+.BR v1.0.7\~ "released December 19th, 2014"
+Fixes a bug in the output from chimera detection with the --uchimeout option.
 .LP
 .\" ============================================================================
 .\" TODO:

diff --git a/doc/vsearch_manual.pdf b/doc/vsearch_manual.pdf
diff --git a/eval/stats.pl b/eval/stats.pl
@@ -40,7 +40,7 @@
 close I;
 
 my $pos = $truepos + $falsepos;
-my $falseneg = $gold - $ truepos;
+my $falseneg = $gold - $truepos;
 my $prec = 100.0 * $truepos / $pos;
 my $fdr  = 100.0 * $falsepos / $pos;
 my $recall = 100.0 * $truepos / $gold;

diff --git a/src/align_simd.cc b/src/align_simd.cc
@@ -732,7 +732,7 @@ void search16(s16info_s * s,
 
   __m128i M_QR_target_left, M_R_target_left;
 
-  __m128i QR_query_left, R_query_left;
+  __m128i R_query_left;
   __m128i QR_query_interior, R_query_interior;
   __m128i QR_query_right, R_query_right;
   __m128i QR_target_left, R_target_left;
@@ -760,8 +760,6 @@ void search16(s16info_s * s,
 
   T0 = _mm_set_epi16(0, 0, 0, 0, 0, 0, 0, 0xffff);
 
-  QR_query_left = _mm_set1_epi16(s->penalty_gap_open_query_left + 
-                                 s->penalty_gap_extension_query_left);
   R_query_left = _mm_set1_epi16(s->penalty_gap_extension_query_left);
 
   QR_query_interior = _mm_set1_epi16(s->penalty_gap_open_query_interior + 

diff --git a/src/chimera.cc b/src/chimera.cc
@@ -337,6 +337,8 @@ int find_best_parents(struct chimera_info_s * ci)
 
 int eval_parents(struct chimera_info_s * ci)
 {
+  int status = 1;
+
   /* create msa */
 
   /* find max insertions in front of each position in the query sequence */
@@ -622,11 +624,10 @@ int eval_parents(struct chimera_info_s * ci)
       }
   }
 
-  int ischimeric = 0;
-  int isnonchimeric = 0;
-
   if (best_h >= 0.0)
     {
+      status = 2;
+
       /* flip A and B if necessary */
 
       if (best_reverse)
@@ -733,19 +734,18 @@ int eval_parents(struct chimera_info_s * ci)
 
       if (best_h >= opt_minh)
         {
+          status = 3;
           if ((divdiff >= opt_mindiv) &&
               (sumL >= opt_mindiffs) &&
               (sumR >= opt_mindiffs))
-            ischimeric = 1;
+            status = 4;
         }
-      else
-        isnonchimeric = 1;
 
       /* print alignment */
 
       pthread_mutex_lock(&mutex_output);
 
-      if (opt_uchimealns && ischimeric)
+      if (opt_uchimealns && (status == 4))
         {
           fprintf(fp_uchimealns, "\n");
           fprintf(fp_uchimealns, "----------------------------------------"
@@ -851,7 +851,7 @@ int eval_parents(struct chimera_info_s * ci)
                       best_right_n,
                       best_right_a,
                       divdiff,
-                      ischimeric ? 'Y' : (isnonchimeric ? 'N' : '?'));
+                      status == 4 ? 'Y' : (status == 2 ? 'N' : '?'));
             }
           else
             {
@@ -875,24 +875,24 @@ int eval_parents(struct chimera_info_s * ci)
                       best_right_n,
                       best_right_a,
                       divdiff,
-                      ischimeric ? 'Y' : (isnonchimeric ? 'N' : '?'));
+                      status == 4 ? 'Y' : (status == 2 ? 'N' : '?'));
             }
         }
       pthread_mutex_unlock(&mutex_output);
     }
-  else
-    {
-      isnonchimeric = 1;
-    }
-
-  if (ischimeric)
-    return 1;
-  else if (isnonchimeric)
-    return -1;
-  else
-    return 0;
+
+  return status;
 }
 
+/*
+  new chimeric status:
+  0: no parents, non-chimeric
+  1: score < 0 (no alignment), non-chimeric
+  2: score < minh, non-chimeric
+  3: score >= minh, suspicious
+  4: score >= minh && (divdiff >= opt_mindiv) && ..., chimeric
+*/
+
 void query_init(struct searchinfo_s * si)
 {
   si->qsequence = 0;
@@ -1152,13 +1152,13 @@ unsigned long chimera_thread_core(struct chimera_info_s * ci)
       if (find_best_parents(ci))
         status = eval_parents(ci);
       else
-        status = -1;
+        status = 0;
 
       /* output results */
 
       pthread_mutex_lock(&mutex_output);
 
-      if (status > 0)
+      if (status == 4)
         {
           chimera_count++;
 
@@ -1170,12 +1170,12 @@ unsigned long chimera_thread_core(struct chimera_info_s * ci)
             }
         }
 
-      if (status < 0)
+      if (status < 3)
         {
           nonchimera_count++;
 
           /* output no parents, no chimeras */
-          if (opt_uchimeout)
+          if ((status < 2) && opt_uchimeout)
             {
               if (opt_uchimeout5)
                 fprintf(fp_uchimeout,

diff --git a/src/db.cc b/src/db.cc
@@ -82,7 +82,7 @@ void db_read(const char * filename, int upcase)
   rewind(fp);
 
   char * prompt;
-  asprintf(& prompt, "Reading file %s", filename);
+  (void) asprintf(& prompt, "Reading file %s", filename);
   progress_init(prompt, filesize);
 
 #ifdef HAVE_BZLIB
@@ -124,7 +124,7 @@ void db_read(const char * filename, int upcase)
   switch (db_format)
    {
      case FORMAT_PLAIN:
-       fgets(line, LINEALLOC, fp);
+       (void) fgets(line, LINEALLOC, fp);
        break;
      case FORMAT_BZIP:
 #ifdef HAVE_BZLIB
@@ -205,7 +205,7 @@ void db_read(const char * filename, int upcase)
       switch (db_format)
        {
          case FORMAT_PLAIN:
-           fgets(line, LINEALLOC, fp);
+           (void) fgets(line, LINEALLOC, fp);
            break;
          case FORMAT_BZIP:
 #ifdef HAVE_BZLIB
@@ -294,7 +294,7 @@ void db_read(const char * filename, int upcase)
           switch (db_format)
            {
              case FORMAT_PLAIN:
-               fgets(line, LINEALLOC, fp);
+               (void) fgets(line, LINEALLOC, fp);
                break;
              case FORMAT_BZIP:
 #ifdef HAVE_BZLIB

diff --git a/src/query.cc b/src/query.cc
@@ -84,7 +84,7 @@ static char * FGETS(char * query_line, int size)
   switch (query_format)
    {
      case FORMAT_PLAIN:
-       fgets(query_line, size, query_fp);
+       (void) fgets(query_line, size, query_fp);
        break;
      case FORMAT_BZIP:
 #ifdef HAVE_BZLIB

diff --git a/src/sortbysize.cc b/src/sortbysize.cc
@@ -75,12 +75,12 @@ void sortbysize()
 
   for(int i=0; i<dbsequencecount; i++)
     {
-      unsigned int size = db_getabundance(i);
+      long size = db_getabundance(i);
 
       if((size >= opt_minsize) && (size <= opt_maxsize))
         {
           sortinfo[passed].seqno = i;
-          sortinfo[passed].size = size;
+          sortinfo[passed].size = (unsigned int) size;
           passed++;
         }
       progress_update(i);

diff --git a/src/util.cc b/src/util.cc
@@ -91,7 +91,7 @@ void * xmalloc(size_t size)
 {
   const size_t alignment = 16;
   void * t;
-  posix_memalign(& t, alignment, size);
+  (void) posix_memalign(& t, alignment, size);
 
   if (t==NULL)
     fatal("Unable to allocate enough memory.");

diff --git a/src/vsearch.cc b/src/vsearch.cc
@@ -439,7 +439,7 @@ void args_init(int argc, char **argv)
   opt_maxqsize = INT_MAX;
   opt_maxqt = DBL_MAX;
   opt_maxrejects = -1;
-  opt_maxseqlength = 50000;
+  opt_maxseqlength = 15000;
   opt_maxsize = LONG_MAX;
   opt_maxsizeratio = DBL_MAX;
   opt_maxsl = DBL_MAX;

diff --git a/src/vsearch.h b/src/vsearch.h
@@ -102,7 +102,7 @@
 #define FORMAT_GZIP  3
 
 #define PROG_NAME "vsearch"
-#define PROG_VERSION "v1.0.6"
+#define PROG_VERSION "v1.0.7"
 
 /* options */
 

diff --git a/test/clusterf.sh b/test/clusterf.sh
@@ -2,9 +2,9 @@
 
 P=$1
 
-INPUT=../data/Rfam_9_1.fasta
+#INPUT=../data/Rfam_9_1.fasta
 #INPUT=../data/AF091148.fsa
-#INPUT=../data/BioMarKs50k.fsa
+INPUT=../data/BioMarKs50k.fsa
 
 THREADS=0
 

diff --git a/test/clusters.sh b/test/clusters.sh
@@ -29,7 +29,6 @@ CMD="/usr/bin/time $PROG \
     --id 0.9 \
     --maxaccepts 1 \
     --maxrejects 8 \
-    --sizein \
     --sizeout \
     --centroids s.$P.centroids \
     --uc s.$P.uc \