Merge pull request #4 from refresh-bio/splash2.1.4

Splash2.1.4

Author: marekkokot

Makefile

@@ -1,6 +1,6 @@
-all: satc satc_dump satc_merge sig_anch download_kmc nomad
+all: satc satc_dump satc_merge sig_anch download_kmc splash supervised_test
 
-NOMAD_LIBS_DIR = libs
+SPLASH_LIBS_DIR = libs
 LIBS_DIR = . #/usr/local/lib
 INCLUDE_DIR= libs
 MIMALLOC_INLUCDE_DIR = libs/mimalloc/include
@@ -14,14 +14,16 @@ COMMON_DIR=src/common
 OUT_BIN_DIR=bin
 
 CC 	= g++
-CFLAGS	= -fPIC -Wall -O3 -m64 -std=c++17 -mavx -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
-CLINK	= -lm -std=c++17 -lpthread -static-libstdc++ -lgfortran
+CFLAGS	= -fPIC -Wall -O3 -m64 -std=c++17 -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
+CLINK	= -lm -std=c++17 -lpthread -static-libstdc++
 
-release: CLINK	= -lm -std=c++17 -static -lgfortran -lquadmath -Wl,--whole-archive -lpthread -Wl,--no-whole-archive
-release: CFLAGS	= -fPIC -Wall -O3 -DNDEBUG -m64 -std=c++17 -mavx -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
+MIMALLOC_OBJ=libs/mimalloc/mimalloc.o
+
+release: CLINK	= -lm -std=c++17 -static -Wl,--whole-archive -lpthread -Wl,--no-whole-archive
+release: CFLAGS	= -fPIC -Wall -O3 -DNDEBUG -m64 -std=c++17 -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
 release: all
 
-debug: CFLAGS	= -fPIC -Wall -O0 -g -m64 -std=c++17 -mavx -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
+debug: CFLAGS	= -fPIC -Wall -O0 -g -m64 -std=c++17 -pthread -I $(INCLUDE_DIR) -I $(MIMALLOC_INLUCDE_DIR) -fpermissive
 debug: all
 
 ifdef MSVC     # Avoid the MingW/Cygwin sections
@@ -47,18 +49,22 @@ prefix      = /usr/local
 # optional install location
 exec_prefix = $(prefix)
 
+$(MIMALLOC_OBJ):
+	$(CC) -DMI_MALLOC_OVERRIDE -O3 -DNDEBUG -fPIC -Wall -Wextra -Wno-unknown-pragmas -fvisibility=hidden -ftls-model=initial-exec -fno-builtin-malloc -c -I libs/mimalloc/include libs/mimalloc/src/static.c -o $(MIMALLOC_OBJ)
+
 %.o: %.cpp
 	$(CC) $(CFLAGS) -c $< -o $@
 
 satc: $(OUT_BIN_DIR)/satc
 
 $(OUT_BIN_DIR)/satc: $(SATC_MAIN_DIR)/satc.o \
-	$(SATC_MAIN_DIR)/kmc_api/kmc_file.o \
-	$(SATC_MAIN_DIR)/kmc_api/mmer.o \
-	$(SATC_MAIN_DIR)/kmc_api/kmer_api.o
+	$(COMMON_DIR)/kmc_api/kmc_file.o \
+	$(COMMON_DIR)/kmc_api/mmer.o \
+	$(COMMON_DIR)/kmc_api/kmer_api.o \
+	$(COMMON_DIR)/illumina_adapters_static.o
 	-mkdir -p $(OUT_BIN_DIR)
 	$(CC) -o $@ $^ \
-	$(NOMAD_LIBS_DIR)/$(LIB_ZSTD) \
+	$(SPLASH_LIBS_DIR)/$(LIB_ZSTD) \
 	$(CLINK)
 
 satc_merge: $(OUT_BIN_DIR)/satc_merge
@@ -69,15 +75,15 @@ $(OUT_BIN_DIR)/satc_merge: $(SATC_MERGE_MAIN_DIR)/satc_merge.o \
 	$(SATC_MERGE_MAIN_DIR)/extra_stats.o
 	-mkdir -p $(OUT_BIN_DIR)
 	$(CC) -o $@ $^ \
-	$(NOMAD_LIBS_DIR)/$(LIB_ZSTD) \
+	$(SPLASH_LIBS_DIR)/$(LIB_ZSTD) \
 	$(CLINK)
 
 satc_dump: $(OUT_BIN_DIR)/satc_dump
 
 $(OUT_BIN_DIR)/satc_dump: $(SATC_DUMP_MAIN_DIR)/satc_dump.o
 	-mkdir -p $(OUT_BIN_DIR)
 	$(CC) -o $@ $^ \
-	$(NOMAD_LIBS_DIR)/$(LIB_ZSTD) \
+	$(SPLASH_LIBS_DIR)/$(LIB_ZSTD) \
 	$(CLINK)
 
 sig_anch: $(OUT_BIN_DIR)/sig_anch
@@ -92,24 +98,26 @@ download_kmc:
 	-mkdir -p $(OUT_BIN_DIR)
 	./download_kmc.sh $(OUT_BIN_DIR)
 
-nomad:
-	cp src/nomad.py bin/nomad
+splash:
+	cp src/splash.py bin/splash
+supervised_test:
+	cp src/supervised_test/supervised_test.R bin
 
 install: all
 	install bin/* /usr/local/bin
 
 uninstall:
-	-rm /usr/local/bin/satc 
+	-rm /usr/local/bin/satc
 	-rm /usr/local/bin/satc_dump
 	-rm /usr/local/bin/satc_merge
-	-rm /usr/local/bin/sig_anch 
-	-rm /usr/local/bin/nomad
+	-rm /usr/local/bin/sig_anch
+	-rm /usr/local/bin/splash
 	-rm /usr/local/bin/kmc
 	-rm /usr/local/bin/kmc_tools
 
-clean:	
+clean:
 	-rm $(SATC_MAIN_DIR)/*.o
-	-rm $(SATC_MAIN_DIR)/kmc_api/*.o	
+	-rm $(COMMON_DIR)/kmc_api/*.o
 	-rm $(SATC_MERGE_MAIN_DIR)/*.o
 	-rm $(SATC_DUMP_MAIN_DIR)/*.o
 	-rm $(COMMON_DIR)/*.o

NOMAD_extendor_classification.R

@@ -41,7 +41,7 @@ riffle <- function(a, b) {   # this function interleaves the elements of two vec
 ##################################################
 ############ input arguments #####################
 args <- commandArgs(trailingOnly = TRUE)
-directory = args[1]  # the output directory used for the NOMAD run
+directory = args[1]  # the output directory used for the SPLASH run
 which_anchors_file = args[2]  # flag to decide which anchor file (after correction or all anchors) to use, could be "after_correction" or "all" 
 effect_size_cutoff = args[3] # the effect size cutoff for significant anchors (default 0.2) 
 num_samples_cutoff = args[4] # the minimum number of sampels for an anchor to be called (default 20)

README.md

@@ -11,19 +11,21 @@ def replace_in_file(file_path, search_text, new_text):
             new_line = line.replace(search_text, new_text)
             print(new_line, end='')
 
-def get_ver(nomad_path):
-    with open(nomad_path) as f:
+def get_ver(splash_path):
+    with open(splash_path) as f:
         for line in f.readlines():
             line = line.strip()
-            if "NOMAD_VERSION" in line:
+            if "SPLASH_VERSION" in line:
                 return line.split("=")[-1].strip().split("\"")[1]
-    print("Error: cannot read NOMAD_VERSION")
+    print("Error: cannot read SPLASH_VERSION")
     sys.exit(1)
 
 def run_cmd(cmd):    
     p = subprocess.Popen(cmd, shell=True)
     p.communicate()
 
+run_cmd("git submodule init")
+run_cmd("git submodule update")
 run_cmd("make clean")
 run_cmd("make -j32 release")
 
@@ -37,12 +39,12 @@ def run_cmd(cmd):
 with open("bin/example/run-example.sh", "w") as f:
     f.write("#!/bin/bash\n")
     f.write("./download.py\n")
-    f.write("../nomad --bin_path .. input.txt\n")
+    f.write("../splash --bin_path .. input.txt\n")
 
 os.chmod("bin/example/run-example.sh", os.stat("bin/example/run-example.sh").st_mode | stat.S_IEXEC)
 
 
-ver = get_ver("bin/nomad")
+ver = get_ver("bin/splash")
 
-run_cmd(f"cd bin; tar -c * | pigz > ../nomad-{ver}.linux.x64.tar.gz; cd ..;")
+run_cmd(f"cd bin; tar -c * | pigz > ../splash-{ver}.linux.x64.tar.gz; cd ..;")
 run_cmd("rm -rf bin")

build_release.py

@@ -30,7 +30,7 @@
    "metadata": {},
    "outputs": [],
    "source": [
-    "NOMAD2_result_folder=''"
+    "SPLASH2_result_folder=''"
    ]
   },
   {
@@ -41,7 +41,7 @@
    "outputs": [],
    "source": [
     "### read in p-values, afetr_correction.scores.csv file\n",
-    "fname=NOMAD2_result_folder+'/result.after_correction.scores.tsv'\n",
+    "fname=SPLASH2_result_folder+'/result.after_correction.scores.tsv'\n",
     "df = pd.read_csv(fname,sep='\\t')"
    ]
   },
@@ -95,7 +95,7 @@
     "dfArr = []\n",
     "\n",
     "##### path to intermediary_files output folder\n",
-    "fldrName= NOMAD2_result_folder+'/result_Cjs/'\n",
+    "fldrName= SPLASH2_result_folder+'/result_Cjs/'\n",
     "\n",
     "for fname in tqdm(glob.glob(fldrName+'bin*.cjs')):\n",
     "    dfArr.append(\n",

example/analysis_scripts/c_analysis.ipynb

@@ -66,7 +66,7 @@ def get_args():
     parser.add_argument(
         "--satc_dump_file",
         type=str,
-        help='Path to satc_dump utility file (within NOMAD2, /bin/satc_dump)'
+        help='Path to satc_dump utility file (within SPLASH2, /bin/satc_dump)'
     )
 
     args = parser.parse_args()

example/analysis_scripts/plotGeneration.py

@@ -0,0 +1,129 @@
+#pragma once
+#include <random>
+#include <tuple>
+#include <iostream>
+
+#undef min
+#undef max
+
+template<class IntType = int>
+class det_uniform_int_distribution {
+public:
+	// types
+	typedef IntType result_type;
+	typedef std::pair<int, int> param_type;
+
+	// constructors and reset functions
+	explicit det_uniform_int_distribution(IntType a = 0, IntType b = std::numeric_limits<IntType>::max());
+	explicit det_uniform_int_distribution(const param_type& parm);
+	void reset();
+
+	// generating functions
+	template<class URNG>
+	result_type operator()(URNG& g);
+	template<class URNG>
+	result_type operator()(URNG& g, const param_type& parm);
+
+	// property functions
+	result_type a() const;
+	result_type b() const;
+	param_type param() const;
+	void param(const param_type& parm);
+	result_type min() const;
+	result_type max() const;
+
+private:
+	typedef typename std::make_unsigned<IntType>::type diff_type;
+
+	IntType lower;
+	IntType upper;
+};
+
+template<class IntType>
+det_uniform_int_distribution<IntType>::det_uniform_int_distribution(IntType a, IntType b) {
+	lower = a;
+	upper = b;
+}
+
+template<class IntType>
+det_uniform_int_distribution<IntType>::det_uniform_int_distribution(const param_type& parm) {
+	param(parm);
+}
+
+template<class IntType>
+void det_uniform_int_distribution<IntType>::reset() {}
+
+template<class IntType>
+template<class URNG>
+auto det_uniform_int_distribution<IntType>::operator()(URNG& g) -> result_type {
+	return operator()(g, param());
+}
+
+template<class IntType>
+template<class URNG>
+auto det_uniform_int_distribution<IntType>::operator()(URNG& g, const param_type& parm) -> result_type {
+	diff_type diff = (diff_type)parm.second - (diff_type)parm.first + 1;
+	if (diff == 0) // If the +1 overflows we are using the full range, just return g()
+		return g();
+
+	diff_type badDistLimit = std::numeric_limits<diff_type>::max() / diff;
+	do {
+		diff_type generatedRand = g();
+
+		if (generatedRand / diff < badDistLimit)
+			return (IntType)((generatedRand % diff) + (diff_type)parm.first);
+	} while (true);
+}
+
+template<class IntType>
+auto det_uniform_int_distribution<IntType>::a() const -> result_type {
+	return lower;
+}
+
+template<class IntType>
+auto det_uniform_int_distribution<IntType>::b() const -> result_type {
+	return upper;
+}
+
+template<class IntType>
+auto det_uniform_int_distribution<IntType>::param() const -> param_type {
+	return param_type(lower, upper);
+}
+
+template<class IntType>
+void det_uniform_int_distribution<IntType>::param(const param_type& parm) {
+	std::tie(lower, upper) = parm;
+	if (upper < lower)
+		throw std::exception();
+}
+
+template<class IntType>
+auto det_uniform_int_distribution<IntType>::min() const -> result_type {
+	return lower;
+}
+
+template<class IntType>
+auto det_uniform_int_distribution<IntType>::max() const -> result_type {
+	return upper;
+};
+
+
+
+
+template<class RandomIt, class URBG>
+void partial_shuffle(RandomIt first, RandomIt middle, RandomIt last, URBG&& g)
+{
+	typedef typename std::iterator_traits<RandomIt>::difference_type diff_t;
+	typedef det_uniform_int_distribution<diff_t> distr_t;
+	typedef typename distr_t::param_type param_t;
+
+	distr_t D;
+	diff_t n = middle - first;
+	diff_t N = last - first - 1;
+	for (diff_t i =0; i < n; ++i) {
+		using std::swap;
+		swap(first[i], first[D(g, param_t(i, N))]);
+	}
+}
+
+

libs/refresh/deterministic_random.h

@@ -187,6 +187,11 @@ namespace refresh {
 			}
 		}
 
+		bool check_completed() {
+			std::lock_guard<std::mutex> lck(mtx);
+			return is_completed;
+		}
+		
 		void cancel()
 		{
 			std::lock_guard<std::mutex> lck(mtx);