Added new distribution: LogUniform

Makefile

@@ -23,6 +23,7 @@ D_LIBS= -lgsl -lgslcblas -lsqlite3 -lmpfr
 
 #Static libraries for MAC
 _S_LIBS= libgsl.a libgslcblas.a libsqlite3.a libgmp.a libmpfr.a
+LD_LIBRARY_PATH=/opt/local/lib
 MS_LIBS=$(patsubst %,$(LD_LIBRARY_PATH)/%,$(_S_LIBS)) #BSD's LD needs the full path
 
 #Static libraries for Linux

Sim-Phy.xcodeproj/xcuserdata/Diego.xcuserdatad/xcdebugger/Breakpoints_v2.xcbkptlist

@@ -842,11 +842,11 @@
             ignoreCount = "0"
             continueAfterRunningActions = "No"
             filePath = "src/sampling.c"
-            timestampString = "458058871.983623"
+            timestampString = "469147207.671261"
             startingColumnNumber = "9223372036854775807"
             endingColumnNumber = "9223372036854775807"
-            startingLineNumber = "138"
-            endingLineNumber = "138"
+            startingLineNumber = "140"
+            endingLineNumber = "140"
             landmarkName = "sample_distr()"
             landmarkType = "7">
          </BreakpointContent>
@@ -858,11 +858,11 @@
             ignoreCount = "0"
             continueAfterRunningActions = "No"
             filePath = "src/sampling.c"
-            timestampString = "465449327.617127"
+            timestampString = "469147207.671261"
             startingColumnNumber = "9223372036854775807"
             endingColumnNumber = "9223372036854775807"
-            startingLineNumber = "331"
-            endingLineNumber = "331"
+            startingLineNumber = "353"
+            endingLineNumber = "353"
             landmarkName = "Print_Sampling()"
             landmarkType = "7">
          </BreakpointContent>
@@ -954,11 +954,11 @@
             ignoreCount = "0"
             continueAfterRunningActions = "No"
             filePath = "src/sampling.c"
-            timestampString = "465448052.658499"
+            timestampString = "469147207.671261"
             startingColumnNumber = "9223372036854775807"
             endingColumnNumber = "9223372036854775807"
-            startingLineNumber = "265"
-            endingLineNumber = "265"
+            startingLineNumber = "287"
+            endingLineNumber = "287"
             landmarkName = "ParseSampling()"
             landmarkType = "7">
          </BreakpointContent>
@@ -995,5 +995,21 @@
             landmarkType = "7">
          </BreakpointContent>
       </BreakpointProxy>
+      <BreakpointProxy
+         BreakpointExtensionID = "Xcode.Breakpoint.FileBreakpoint">
+         <BreakpointContent
+            shouldBeEnabled = "Yes"
+            ignoreCount = "0"
+            continueAfterRunningActions = "No"
+            filePath = "src/sampling.c"
+            timestampString = "469149337.983703"
+            startingColumnNumber = "9223372036854775807"
+            endingColumnNumber = "9223372036854775807"
+            startingLineNumber = "158"
+            endingLineNumber = "158"
+            landmarkName = "sample_distr()"
+            landmarkType = "7">
+         </BreakpointContent>
+      </BreakpointProxy>
    </Breakpoints>
 </Bucket>

Sim-Phy.xcodeproj/xcuserdata/Diego.xcuserdatad/xcschemes/Sim-Phy.xcscheme

@@ -71,6 +71,10 @@
             argument = "-sb f:0.00001 -sp f:10000"
             isEnabled = "NO">
          </CommandLineArgument>
+         <CommandLineArgument
+            argument = "-i /Users/Diego/Desktop/newsim/SimPhy_configV1.conf"
+            isEnabled = "YES">
+         </CommandLineArgument>
          <CommandLineArgument
             argument = "-i /Users/Diego/Desktop/Sim-Phy_Debug/itrees.txt"
             isEnabled = "NO">

src/main.c

@@ -1512,7 +1512,7 @@ int main (int argc, char **argv)
 
 void PrintUsage(void)
 {
-    printf("\nUsage: ./SimPhy -[Parameter code] value(i|r|c|b|*) ...\n\nValue kinds\n\ti=integer\n\tr=real\n\tc=character string\n\tb=boolean(1 or 0)\n\t*=sampling notation\n\nPARAMETERS\n__________\n__________\n\n-R...→ Replicates\n_________________\n\n\t-RS i: Number of species tree replicates (study replicates).\n\t-RL *: Number of locus trees per species tree.\n\t-RG i: Number of gene trees per locus tree (Not for general usage).\n\n-G... → Genome-wide parameters (sampled for each species tree)\n______________________________________________________________\n\n\t-GB *: Duplication parameter (to use with LB)\n\t-GD *: Loss rate parameter (to use with LD).\n\t-GT *: Transfer parameter (to use with LT).\n\t-GG *: Gene conversion parameter (to use with LG).\n\t-GP *: Gene-by-lineage-specific parameter (to use with HG).\n\n-S... → Species tree\n____________________\n\n\t-S c: Fixed species tree (extended Newick format).\n\t-SR c: Nexus file with species trees.\n\t-SB *: Speciation rate (events/time unit).\n\t-SD *: Extinction rate (events/time unit).\n\t-ST *: Species tree height (time units).\n\t-SL *: Number of taxa.\n\t-SO *: Ratio between ingroup height and the branch from the root to the ingroup. If this parameter is not set the outgroup is not simulated.\n\t-SI *: Number of individuals per species.\n\t-SP *: Tree-wide effective population size.\n\t-SU *: Tree-wide substitution rate.\n\t-SG *: Tree-wide generation time.\n\n-L... → Locus tree\n__________________\n\n\t-L c: Locus tree (extended Newick format).\n\t-LR c: Nexus file containing locus trees.\n\t-LB *: Duplication rate (events/generation).\n\t-LD *: Loss rate (events/generation).\n\t-LT *: Horizontal gene transfer (HGT) rate (events/generation). \n\t-LG *: Gene conversion (GC) rate (events/generation).\n\t-LK b: Distance-dependent HGT/GC: Determines whether the sampling of receptors of genetic material depends (1) or not (0) on the evolutionary distance between candidates and donors.\n\t-LL i: Minimum number of locus tree leaves.\n\t-LS i: Minimum number of species represented by the locus tree leaves.\n\n-H...→ Substitution rate heterogeneity parameters\n_________________________________________________\n\n\t-HS *: Species-specific branch rate heterogeneity modifiers.\n\t-HL *: Gene-family-specific rate heterogeneity modifiers.\n\t-HH *:Gene-by-lineage-specific locus tree parameter (to use with the HG argument below).\n\t-HG *: Gene-by-lineage-specific rate heterogeneity modifiers.\n\n-C... → Global options\n______________________\n\n\t-CS i: Random number generator seed.\n\t-CE r: Precision of the Brent’s method for root-finding when sampling the multilocus coalescent. (Not for general usage)\n\n-I c: Input configuration file\n\n-V [0,6]: Verbosity. (Note: the bigger the verbosity the slower SimPhy becomes. Levels over 3 may only make sense for debugging or study the way SimPhy works).\n\n-O...→ Output\n_____________\n\n\t-O c: Common output prefix-name (for folder and names).\n\t-OT b: Determines whether the species and locus tree branches are written in number of generations (0) or time units (1).\n\t-OM b: Activates the tree mapping output.\n\t-OD b: Activates the SQLite database output.\n\t-OP b: Activates the logging of sampled options.\n\t-OC b: Activates the logging of original command line parameters and input configuration files.\n\t-OL b: Activates the output of trees with internal nodes labelled by its post-order id starting from 0.\n\t-ON b: Activates the output of the bounded locus subtrees file.\n\n\nSampling notation\n_________________\n\nNotation squeme= Distribution_code:parameter_1,parameter_2,...,parameter_n\nDistribution codes:\n\tF: fixed value\n\tU: Uniform\n\tN: Normal\n\tE: Exponential\n\tG: Gamma\n\tL: Lognormal\n\tSL: Lognormal * constant\nExample: N:1,1 (Normal with mean and sd equal 1)\n\nExample\n_______\n\nsimphy -sb f:0.000001 -ld f:0.0000005 -lb f:0.0000005 -lt f:0.0000005 -rs 100 -rl U:10,100 -rg 1 -o SimPhy_test -sp f:10000 -su f:0.00001 -sg f:1 -sl U:20,50 -st f:1000000 -om 1 -v 2 -od 1 -op 1 -oc 1 -on 1 -cs 22\n\n");
+    printf("\nUsage: ./SimPhy -[Parameter code] value(i|r|c|b|*) ...\n\nValue kinds\n\ti=integer\n\tr=real\n\tc=character string\n\tb=boolean(1 or 0)\n\t*=sampling notation\n\nPARAMETERS\n__________\n__________\n\n-R...→ Replicates\n_________________\n\n\t-RS i: Number of species tree replicates (study replicates).\n\t-RL *: Number of locus trees per species tree.\n\t-RG i: Number of gene trees per locus tree (Not for general usage).\n\n-G... → Genome-wide parameters (sampled for each species tree)\n______________________________________________________________\n\n\t-GB *: Duplication parameter (to use with LB)\n\t-GD *: Loss rate parameter (to use with LD).\n\t-GT *: Transfer parameter (to use with LT).\n\t-GG *: Gene conversion parameter (to use with LG).\n\t-GP *: Gene-by-lineage-specific parameter (to use with HG).\n\n-S... → Species tree\n____________________\n\n\t-S c: Fixed species tree (extended Newick format).\n\t-SR c: Nexus file with species trees.\n\t-SB *: Speciation rate (events/time unit).\n\t-SD *: Extinction rate (events/time unit).\n\t-ST *: Species tree height (time units).\n\t-SL *: Number of taxa.\n\t-SO *: Ratio between ingroup height and the branch from the root to the ingroup. If this parameter is not set the outgroup is not simulated.\n\t-SI *: Number of individuals per species.\n\t-SP *: Tree-wide effective population size.\n\t-SU *: Tree-wide substitution rate.\n\t-SG *: Tree-wide generation time.\n\n-L... → Locus tree\n__________________\n\n\t-L c: Locus tree (extended Newick format).\n\t-LR c: Nexus file containing locus trees.\n\t-LB *: Duplication rate (events/generation).\n\t-LD *: Loss rate (events/generation).\n\t-LT *: Horizontal gene transfer (HGT) rate (events/generation). \n\t-LG *: Gene conversion (GC) rate (events/generation).\n\t-LK b: Distance-dependent HGT/GC: Determines whether the sampling of receptors of genetic material depends (1) or not (0) on the evolutionary distance between candidates and donors.\n\t-LL i: Minimum number of locus tree leaves.\n\t-LS i: Minimum number of species represented by the locus tree leaves.\n\n-H...→ Substitution rate heterogeneity parameters\n_________________________________________________\n\n\t-HS *: Species-specific branch rate heterogeneity modifiers.\n\t-HL *: Gene-family-specific rate heterogeneity modifiers.\n\t-HH *:Gene-by-lineage-specific locus tree parameter (to use with the HG argument below).\n\t-HG *: Gene-by-lineage-specific rate heterogeneity modifiers.\n\n-C... → Global options\n______________________\n\n\t-CS i: Random number generator seed.\n\t-CE r: Precision of the Brent’s method for root-finding when sampling the multilocus coalescent. (Not for general usage)\n\n-I c: Input configuration file\n\n-V [0,6]: Verbosity. (Note: the bigger the verbosity the slower SimPhy becomes. Levels over 3 may only make sense for debugging or study the way SimPhy works).\n\n-O...→ Output\n_____________\n\n\t-O c: Common output prefix-name (for folder and names).\n\t-OT b: Determines whether the species and locus tree branches are written in number of generations (0) or time units (1).\n\t-OM b: Activates the tree mapping output.\n\t-OD b: Activates the SQLite database output.\n\t-OP b: Activates the logging of sampled options.\n\t-OC b: Activates the logging of original command line parameters and input configuration files.\n\t-OL b: Activates the output of trees with internal nodes labelled by its post-order id starting from 0.\n\t-ON b: Activates the output of the bounded locus subtrees file.\n\n\nSampling notation\n_________________\n\nNotation squeme= Distribution_code:parameter_1,parameter_2,...,parameter_n\nDistribution codes:\n\tF: fixed value\n\tU: Uniform\n\tN: Normal\n\tE: Exponential\n\tG: Gamma\n\tLN: Lognormal\n\tLU: LogUniform\n\tSL: Lognormal * constant\nExample: N:1,1 (Normal with mean and sd equal 1)\n\nExample\n_______\n\nsimphy -sb f:0.000001 -ld f:0.0000005 -lb f:0.0000005 -lt f:0.0000005 -rs 100 -rl U:10,100 -rg 1 -o SimPhy_test -sp f:10000 -su f:0.00001 -sg f:1 -sl U:20,50 -st f:1000000 -om 1 -v 2 -od 1 -op 1 -oc 1 -on 1 -cs 22\n\n");
 
     fflush(stdout);
 

src/sampling.c

@@ -107,6 +107,8 @@ long int sample_distr(gsl_rng *r,int n_arg,...)
 {
     va_list ap;
     int i;
+    double a=0;
+    double b=0;
     sampling_unit *variable,**variables;
     variables=calloc(n_arg, sizeof(sampling_unit*));
     va_start(ap, n_arg);
@@ -150,6 +152,11 @@ long int sample_distr(gsl_rng *r,int n_arg,...)
             case LOGNORMAL:
                 set_propsampling(variable, gsl_ran_lognormal(r,GetCastedDoubleUSU(variable->params[0], variable->params_type[0]),GetCastedDoubleUSU(variable->params[1], variable->params_type[1])));
                 break;
+            case LOGUNIFORM:
+                a=log(GetCastedDoubleUSU(variable->params[0], variable->params_type[0]));
+                b=log(GetCastedDoubleUSU(variable->params[1], variable->params_type[1]));
+                set_propsampling(variable, exp(a+(b-a)*gsl_rng_uniform(r)));
+                break;
             case LOGNORMAL_MULT:
                 set_propsampling(variable,(gsl_ran_lognormal(r,GetCastedDoubleUSU(variable->params[0], variable->params_type[0]),GetCastedDoubleUSU(variable->params[1], variable->params_type[1]))*GetCastedDoubleUSU(variable->params[2], variable->params_type[2])));
                 break;
@@ -195,8 +202,23 @@ long int ParseSampling(char * p, sampling_unit * sample, const sampling_table sa
             n_p=2;
             break;
         case 'L':
-            sample->distribution_code=LOGNORMAL;
-            n_p=2;
+            switch (toupper(*(p+1)))
+            {
+                case 'N':
+                    sample->distribution_code=LOGNORMAL;
+                    n_p=2;
+                    ++p;
+                    break;
+                case 'U':
+                    sample->distribution_code=LOGUNIFORM;
+                    n_p=2;
+                    ++p;
+                    break;
+
+                default:
+                    return SETTINGS_ERROR;
+                    break;
+            }
             break;
         case 'S':
             switch (toupper(*(p+1)))
@@ -355,6 +377,10 @@ void Print_Sampling(sampling_unit *sample, char * buffer, const sampling_table s
             sprintf(buffer,"LogN(");
             n_param=2;
             break;
+        case LOGUNIFORM:
+            sprintf(buffer,"LogU[");
+            n_param=2;
+            break;
         case LOGNORMAL_MULT:
             sprintf(buffer,"LogN(");
             n_param=3;

src/sampling.h

@@ -56,7 +56,7 @@ extern int MAX_IT;
  *  Normal distribution (2 parameters).
  *******************************************************************************/
 
-enum DISTRIBUTIONS {FIXED=0, UNIFORM=1, NORMAL=2, EXPONENTIAL=3, GAMMA=4, LOGNORMAL=5, LOGNORMAL_MULT=7};
+enum DISTRIBUTIONS {FIXED=0, UNIFORM=1, NORMAL=2, EXPONENTIAL=3, GAMMA=4, LOGNORMAL=5,LOGUNIFORM=6, LOGNORMAL_MULT=7};
 
 /**
  * \enum TYPES