Merge pull request #7 from mrc-ide/develop

First draft of Tapestry including a basic IBD model and inference via PT-MCMC.

.github/workflows/build.yml

@@ -0,0 +1,42 @@
+name: Build-Test
+on: [push, pull_request]
+env:
+  # Customize the CMake build type here (Release, Debug, RelWithDebInfo, etc.)
+  BUILD_TYPE: Release
+jobs:
+  build:
+    runs-on: ubuntu-latest
+    steps:
+      - name: Checkout Tapestry
+        uses: actions/checkout@v3
+
+      - name: Checkout HTSlib
+        uses: actions/checkout@v3
+        with:
+          submodules: true  # HTSlib uses 'htscodecs' as a submodule
+          repository: samtools/htslib
+          path: htslib
+
+      - name: Build and install HTSlib
+        run: |
+          cd htslib
+          autoreconf -i
+          ./configure --prefix ${{github.workspace}}/htslib
+          make
+          make install
+
+      - name: Configure CMake
+        # Configure CMake in a 'build' subdirectory. `CMAKE_BUILD_TYPE` is only required if you are using a single-configuration generator such as make.
+        # See https://cmake.org/cmake/help/latest/variable/CMAKE_BUILD_TYPE.html?highlight=cmake_build_type
+        run: cmake -B ${{github.workspace}}/build -DCMAKE_BUILD_TYPE=${{env.BUILD_TYPE}}
+
+      - name: Build
+        # Build your program with the given configuration
+        run: cmake --build ${{github.workspace}}/build --config ${{env.BUILD_TYPE}} -DHTSLIB_DIR=${{ github.workspace }}/htslib
+
+      - name: Test
+        working-directory: ${{github.workspace}}/build
+        # Execute tests defined by the CMake configuration.
+        # See https://cmake.org/cmake/help/latest/manual/ctest.1.html for more detail
+        run: ctest -C ${{env.BUILD_TYPE}} -VV
+

.gitignore

@@ -1,5 +1,27 @@
-.Rproj.user
-.Rhistory
-.RData
-.Ruserdata
-ignore/
+# macOS
+.DS_Store
+
+# VSCode
+.vscode
+
+# Compilation
+release/
+build/
+!build/cmake/modules/*
+
+# Python
+__pycache__
+notebooks
+*.ipynb
+*.ipynb_checkpoints
+*.egg
+*.egg-info
+
+# Run outputs
+results
+
+# Personal
+src/dev*
+dev*
+images/*.ai
+TODO.txt

CMakeLists.txt

@@ -0,0 +1,105 @@
+# ================================================================================
+# Project Informationn
+#
+# ===============================================================================
+
+cmake_minimum_required(VERSION 3.14)
+project(apropos VERSION 0.0
+                DESCRIPTION "Tapestry: Inference from complex P.f. infections"
+                LANGUAGES CXX
+)
+
+set(CMAKE_RUNTIME_OUTPUT_DIRECTORY ${CMAKE_BINARY_DIR})
+set(CMAKE_MODULE_PATH ${CMAKE_MODULE_PATH} "${CMAKE_SOURCE_DIR}/build/cmake/modules/")
+
+# Compilation flags
+set(COMPILE_TESTS ON)
+
+
+# ================================================================================
+# Use C++17
+#
+# ================================================================================
+
+set(CMAKE_CXX_STANDARD 17)              # required for <filesystem>
+set(CMAKE_CXX_STANDARD_REQUIRED ON)
+set(CMAKE_CXX_EXTENSIONS OFF)
+
+# ================================================================================
+# Install GoogleTest
+#
+# ================================================================================
+
+if (COMPILE_TESTS)
+    include(FetchContent)
+    FetchContent_Declare(
+        googletest
+        URL https://github.com/google/googletest/archive/5376968f6948923e2411081fd9372e71a59d8e77.zip
+    )
+    set(gtest_force_shared_crt ON CACHE BOOL "" FORCE)
+    FetchContent_MakeAvailable(googletest)
+endif()
+
+# ================================================================================
+# Compile `tapestry`
+#
+# ================================================================================
+
+# Target executable
+set(TARGET tapestry)
+
+# Source files
+set(SOURCES 
+    src/bed.cpp
+    src/betabin.cpp
+    src/combinatorics.cpp
+    src/data.cpp
+    src/ibd.cpp
+    src/io.cpp
+    src/mcmcs.cpp
+    src/model_compare.cpp
+    src/model_fitting.cpp
+    src/models.cpp
+    src/parameters.cpp
+    src/particle_writers.cpp
+    src/particles.cpp
+    src/proposals.cpp
+    src/sampling.cpp
+    src/vcf.cpp
+    )
+
+# Add the executable
+add_executable(${TARGET} src/main.cpp ${SOURCES})
+
+# --------------------------------------------------------------------------------
+# Link htslib
+# --------------------------------------------------------------------------------
+
+# Define htslib_FOUND, htslib_INCLUDE_DIRS, htslib_LIBRARIES
+find_package(HTSlib 1.4 REQUIRED)
+
+# Link include directories and library to the executable
+target_include_directories(${TARGET} PUBLIC ${HTSlib_INCLUDE_DIRS})
+target_link_libraries(${TARGET} PUBLIC ${HTSlib_LIBRARIES})
+
+
+# ================================================================================
+# Compile tests
+#
+# ================================================================================
+
+if (COMPILE_TESTS)
+    enable_testing()
+    set(TSOURCES 
+        tests/test_combinatorics.cpp
+        tests/test_particle.cpp
+        tests/test_random.cpp
+        )
+    add_executable(test_tapestry ${TSOURCES} ${SOURCES})
+
+    target_include_directories(test_tapestry PUBLIC ${HTSlib_INCLUDE_DIRS})
+    target_link_libraries(test_tapestry GTest::gtest_main ${HTSlib_LIBRARIES})
+
+    include(GoogleTest)
+    gtest_discover_tests(test_tapestry)
+endif()

DESCRIPTION

@@ -1,26 +1,20 @@
-Package: Tapestry
-Type: Package
-Title: Software for deconvoluting unphased polyclonal genotypes
-Version: 0.1.0
-Authors@R: as.person(c(
-    "Jason Hendry <[email protected]> [aut, cre]", 
-    "Bob Verity <[email protected]> [aut]"
-  ))
-Description: Malaria infections often contain multiple genotypes. When
-    sequenced together, these produce a complex signal that is a mixture of the
-    individual genotypes. Like earlier programs like DEploid, Tapestry
-    attempts to pull these individual genotypes apart by exploiting allele
-    frequency imbalances within a sample. The difference in Tapestry is that we
-    assume a model of allele frequencies rather than a reference panel. We also
-    implement parallel tempered MCMC to ensure good mixing, and produce a
-    different set of outputs.
-License: MIT + file LICENSE
-Encoding: UTF-8
-LazyData: true
-RoxygenNote: 7.1.1
-LinkingTo: 
-    Rcpp
-Imports: 
-    Rcpp
-SystemRequirements: C++11
-BugReports: https://github.com/mrc-ide/tapestry/issues
+Package: Tapestry
+Type: Package
+Title: Software for deconvoluting unphased polyclonal genotypes
+Version: 0.1.0
+Authors@R: as.person(c(
+    "Jason Hendry <[email protected]> [aut, cre]", 
+    "Bob Verity <[email protected]> [aut]"
+  ))
+Description: Malaria infections often contain multiple genotypes, and when 
+    sequenced together these produce a complex signal that is a mixture of the 
+    individual genotypes. Building on the framework of earlier programs like 
+    DEploid and DEploidIBD, Tapestry attempts to pull these individual genotypes
+    apart by exploiting allele frequency imbalances within a sample, while 
+    simultaneously estimating segments of identity by descent (IBD) between 
+    sequences. Unlike previous programs, Tapestry uses advanced MCMC methods to 
+    ensure that results are robust even for high complexity of infection (COI).
+License: MIT + file LICENSE
+Encoding: UTF-8
+SystemRequirements: C++17
+BugReports: https://github.com/mrc-ide/tapestry/issues