New features, bug fixes, and new test case framework for 4.16.0

.dockerignore

@@ -0,0 +1,8 @@
+cget
+venv
+.git
+private
+build
+clion-build-debug
+clion-build-release
+docker-compose.override.yml

.travis.yml

@@ -0,0 +1,41 @@
+language: cpp
+
+jobs:
+  include:
+    - os: linux
+      dist: bionic
+      compiler: gcc
+      env: LC_ALL=en_US.UTF-8 LANG=en_US.UTF-8
+    - os: osx
+      osx_image: xcode11.3
+      compiler: clang
+
+cache: pip
+
+before_install:
+  - |
+    if [[ "${TRAVIS_OS_NAME}" == "osx" ]]; then
+      brew install cmake zlib xz python3
+      brew cask install gfortran
+      sudo python3 -m pip install cget
+      sudo installer -pkg /Library/Developer/CommandLineTools/Packages/macOS_SDK_headers_for_macOS_10.14.pkg -target /
+      cget install -f requirements.txt
+    fi
+  - |
+    if [[ "${TRAVIS_OS_NAME}" == "linux" ]]; then
+      sudo apt update
+      sudo apt install build-essential cmake gfortran locales zlib1g-dev liblzma-dev
+      sudo apt install python3 python3-setuptools python3-pip python3-virtualenv
+      sudo locale-gen en_US.UTF-8
+      sudo pip3 install cget
+      cget install -f requirements.txt
+    fi
+
+install:
+  - rm -rf build && mkdir build && cd build
+  - cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_TOOLCHAIN_FILE=../cget/cget/cget.cmake -DBUILD_TESTS=1 ..
+  - make
+  - cd ..
+
+script:
+  - build/raremetal/tests/bin/tests-raremetal

CMakeLists.txt

@@ -11,6 +11,8 @@ find_library(LIB_HTS hts)
 find_library(LIB_Z z)
 find_library(LIB_RMATH Rmath)
 
+set(CMAKE_RUNTIME_OUTPUT_DIRECTORY bin)
+
 add_library(libraremetal
             libRareMetal/mvtnorm/randomF77.c
             libRareMetal/pdf/PDF.cpp
@@ -136,8 +138,8 @@ target_include_directories(raremetalworker PRIVATE otherLib/eigen-3.2.0)
 target_link_libraries(raremetalworker libraremetal ${LIB_SAVVY} ${LIB_ZSTD})
 target_compile_definitions(raremetalworker PRIVATE VERSION="${PROJECT_VERSION}")
 
-
 if(BUILD_TESTS)
     enable_testing()
     add_subdirectory(tests)
+    add_subdirectory(raremetal/tests)
 endif()

Dockerfile

@@ -0,0 +1,70 @@
+FROM ubuntu:18.04
+
+ARG BUILD_DATE
+ARG GIT_SHA
+ARG RAREMETAL_VERSION
+
+LABEL org.label-schema.name="RAREMETAL"
+LABEL org.label-schema.description="Meta-analysis of rare genetic variant aggregation tests"
+LABEL org.label-schema.vendor="University of Michigan, Center for Statistical Genetics"
+LABEL org.label-schema.url="https://github.com/statgen/RAREMETAL"
+LABEL org.label-schema.vcs-url="https://github.com/statgen/RAREMETAL"
+LABEL org.label-schema.schema-version="1.0"
+
+# Install required packages for swiss to install. Many of swiss' dependencies
+# require compiling C/C++ code.
+RUN apt-get update && apt-get install -y \
+  build-essential \
+  cmake \
+  gfortran \
+  python3 \
+  python3-pip \
+  zlib1g-dev \
+  liblzma-dev \
+  locales \
+  && rm -rf /var/lib/apt/lists/* \
+  && locale-gen en_US.UTF-8
+
+ENV LC_ALL en_US.UTF-8
+ENV LANG en_US.UTF-8
+
+# Install necessary python packages (backports.lzma needed for cget to extract .xz archives)
+RUN pip3 install cget
+
+# Create a group and user to execute as, then drop root
+ARG UID
+ARG GID
+RUN \
+  if [ -n "$GID" ]; then \
+    addgroup --gid $GID raremetal; \
+  else \
+    addgroup raremetal; \
+  fi && \
+  if [ -n "$UID" ]; then \
+    adduser --gecos "User for running raremetal as non-root" --shell /bin/bash --disabled-password --uid $UID --ingroup raremetal raremetal; \
+  else \
+    adduser --gecos "User for running raremetal as non-root" --shell /bin/bash --disabled-password --ingroup raremetal raremetal; \
+  fi
+
+WORKDIR /home/raremetal
+USER raremetal
+
+# Copy source
+COPY --chown=raremetal:raremetal . /home/raremetal/
+
+# Compile & run tests
+# We want the docker build to fail if the tests fail
+RUN cget install -f requirements.txt
+RUN mkdir build && \
+  cd build && \
+  cmake -DCMAKE_BUILD_TYPE=Release -DCMAKE_TOOLCHAIN_FILE=../cget/cget/cget.cmake -DBUILD_TESTS=1 .. && \
+  make && \
+  cd .. && \
+  build/raremetal/tests/bin/tests-raremetal
+
+ENTRYPOINT ["build/bin/raremetal"]
+
+# Frequently changing metadata here to avoid cache misses
+LABEL org.label-schema.version=$RAREMETAL_VERSION
+LABEL org.label-schema.vcs-ref=$GIT_SHA
+LABEL org.label-schema.build-date=$BUILD_DATE

raremetal/src/Calculate_mvt_pvalue.h

@@ -6,6 +6,8 @@
 #include <stdio.h>
 #include "MathMatrix.h"
 #include "StringHash.h"
+#include "IntArray.h"
+#include "MathSVD.h"
 
 #define MATHLIB_STANDALONE
 
@@ -112,15 +114,9 @@ double CalculateMVTPvalue(Vector &score, Matrix &cov, double t_max)
                     //A new combination has been generated!
                     comb.Push(alpha[indx[owk]]);
                 }
-/*
-            printf("New combination:\n");
-            for(int i=0;i<comb.Length();i++)
-               printf("%g ",comb[i]);
-            printf("\n");
-*/
+
                 //Calculate P(A|BC)
                 GetConditionalDist(score, cov, comb, par);
-//            printf("pars are:%g %g\n",par[0],par[1]);
                 double chisq, condProb, prob;
                 if (par[1] == 0.0)
                 {
@@ -134,7 +130,6 @@ double CalculateMVTPvalue(Vector &score, Matrix &cov, double t_max)
                     }
                     condProb = pchisq(chisq, 1, 0, 0);
                 }
-//          printf("conditional prob is:%g\n",condProb);
                 String hashKey;
 
                 if (r == 2)
@@ -143,10 +138,7 @@ double CalculateMVTPvalue(Vector &score, Matrix &cov, double t_max)
                     hashKey += tmp;
                     tmp = comb[1];
                     hashKey += tmp;
-                    //              printf("hashKey is: %s\n",hashKey.c_str());
                     prob = condProb * uni;
-                    //prob = exp(log(condProb) + log(uni));
-                    //printf("joint prob is:%g\n",prob);
                     jointProbHash.SetDouble(hashKey, prob);
                     hashKey.Clear();
                 } else
@@ -156,13 +148,7 @@ double CalculateMVTPvalue(Vector &score, Matrix &cov, double t_max)
                         int tmp = comb[i];
                         hashKey += tmp;
                     }
-                    //              printf("hashKey is: %s\n",hashKey.c_str());
                     prob = jointProbHash.Double(hashKey);
-//	       printf("joint prob is:%g\n",prob);
-                    //if(prob == _NAN_)
-                    //{
-                    //         printf("Error! Joint probability %s is not hashed in properly.\n",hashKey);
-                    //     }
                     prob *= condProb; //This is the new joint prob.
                     String newKey;
                     int tmp = comb[0];
@@ -174,7 +160,6 @@ double CalculateMVTPvalue(Vector &score, Matrix &cov, double t_max)
                 }
 
                 pvalue -= prob;
-//	    printf("pvalue so far is:%g\n",pvalue);
                 comb.Clear();
                 for (int iwk = r - 1; iwk >= 0; iwk--)
                 {

raremetal/src/GroupFromAnnotation.cpp

@@ -5,15 +5,14 @@
 #include "InputFile.h"
 #include "QuickIndex.h"
 
-String GroupFromAnnotation::groupFile = "";
-String GroupFromAnnotation::vcfInput = "";
-String GroupFromAnnotation::function = "";
-String GroupFromAnnotation::mapFile = "../data/refFlat_hg19.txt";
-bool GroupFromAnnotation::labelHits = false;
-
 GroupFromAnnotation::GroupFromAnnotation()
 {
-    geneCount = 0;
+  geneCount = 0;
+  groupFile = "";
+  vcfInput = "";
+  function = "";
+  mapFile = "../data/refFlat_hg19.txt";
+  labelHits = false;
 }
 
 GroupFromAnnotation::~GroupFromAnnotation()

raremetal/src/GroupFromAnnotation.h

@@ -12,16 +12,14 @@ class GroupFromAnnotation
 {
 public:
     GroupFromAnnotation();
-
     ~GroupFromAnnotation();
 
     //Input/Output options
-    static String vcfInput;
-    static String groupFile;
-    static String function;
-    static String mapFile;
-    static bool labelHits;
-
+    String vcfInput;
+    String groupFile;
+    String function;
+    String mapFile;
+    bool labelHits;
 
     StringArray *SNPlist;
     StringArray *SNPNoAllele;
@@ -31,14 +29,14 @@ class GroupFromAnnotation
     StringIntHash groupHash;
     Vector *pos;
     int geneCount;
+
     //these are for annotating single variant accoding to refFlat_hg19.txt
     IntArray start_pos, end_pos;
     StringArray genename, chr;
     StringIntHash ChrStartHash, ChrEndHash;
     QuickIndex chr_idx;
 
     //this is the position of each marker in a gene for output
-
     void GetGeneMap(String path);
 
     void GetGroupFromFile(FILE *log);
@@ -49,7 +47,7 @@ class GroupFromAnnotation
 
     String AnnotateSingleVar(String chr, int pos);
 
-// group vcf related
+    // group vcf related
     void vcfInitialize();
 
     void addLineFromVcf(String &buffer);