code cleanup & small changes in stats calc

.github/workflows/docker.yml

@@ -32,7 +32,7 @@ jobs:
       - name: extract version
         id: extract_version
         run: | 
-          VERSION=$(grep 'ARG VERSION=' Dockerfile | cut -d'=' -f2)
+          VERSION=${GITHUB_REF#refs/tags/}
           echo "::set-output name=VERSION::$VERSION"
 
       - name: build and push

CHANGELOG.md

@@ -5,6 +5,18 @@ All notable changes to this project will be documented in this file.
 The format is based on [Keep a Changelog](https://keepachangelog.com/en/1.0.0/),
 and this project adheres to [Semantic Versioning](https://semver.org/spec/v2.0.0.html).
 
+# [0.2.2]
+
+## Features
+
+- add test data
+
+## Fix
+
+- Code cleanup
+- Fix in writing to stats.txt that cause overwriting in different subcalls
+
+
 # [0.2.1]
 
 ## Features

CMakeLists.txt

@@ -1,5 +1,5 @@
 cmake_minimum_required(VERSION 3.22.1)
-project(RNAnue VERSION 0.2.1)
+project(RNAnue VERSION 0.2.2)
 set(CMAKE_CXX_STANDARD 20)
 set(CMAKE_CXX_STANDARD_REQUIRED True)
 set(CMAKE_CXX_FLAGS -fopenmp)

README.md

@@ -1,6 +1,6 @@
 [![docker-release](https://github.com/Ibvt/RNAnue/actions/workflows/docker.yml/badge.svg)](https://github.com/Ibvt/RNAnue/actions/workflows/docker.yml)
 
-# RNAnue - 0.2.1
+# RNAnue - 0.2.2
 
 ## About
 RNAnue is a comprehensive analysis to detect RNA-RNA interactions from Direct-Duplex-Detection (DDD) data.
@@ -126,9 +126,8 @@ columns are defined in the following:
 
 ### Interaction table
 
-The `analysis` procedure generates `_interactions` files for each library in 
-which each line represents an annotated split read that is mapped to a 
-transcript interaction. The fields are defined as follows:
+The `analysis` procedure generates `_interactions` files for each library in which each line represents an annotated 
+split read that is mapped to a transcript interaction. The fields are defined as follows:
 
 | Field | Description |
 | ----- | ----------- |
@@ -157,11 +156,9 @@ transcript interaction. The fields are defined as follows:
 | mfe | Hybridisation energy of the interaction |
 | mfe_struc | Minimum free energy (MFE) structure of interaction in dot-bracket notation |
 
-The main result of an RNAnue analysis are transcript interactions. 
-They are stored in  the file `allints.txt` in the same directory. 
-Its entries are structured as described in the following where 
-columns with prefix <sample> are given for each sample specified in 
-the analysis (within the same file).
+The main result of an RNAnue analysis are transcript interactions. They are stored in  the file `allints.txt` in the 
+same directory. Its entries are structured as described in the following where columns with prefix <sample> are given 
+for each sample specified in the analysis (within the same file).
 
 | Field                 | Description |
 |-----------------------| ----------- |
@@ -182,10 +179,12 @@ in JSON graph format. Finally, –stats set to 1 creates a `stats.txt` file that
 each step of the analysis. 
 
 ### Docker
-In additon, we provide a ready-to-use Docker container that has RNAnue preconfigured.
-https://hub.docker.com/repository/docker/cobirna/rnanue
+In additon, we provide a ready-to-use [Docker container](https://hub.docker.com/repository/docker/cobirna/rnanue) that 
+has RNAnue preconfigured.
 
 ### Testing
 
+We provide a test dataset in the [test](./test/data/) folder that can be used to test the installation. 
+
 # Troubleshooting
 contact [email protected] or create an issue

include/Analysis.hpp

@@ -13,24 +13,27 @@
 #include <boost/program_options.hpp>
 #include <boost/filesystem.hpp>
 #include <boost/property_tree/ptree.hpp>
+#include <boost/property_tree/json_parser.hpp>
 #include <boost/accumulators/accumulators.hpp>
 #include <boost/accumulators/statistics.hpp>
 #include <boost/math/distributions/binomial.hpp>
-
+#include <boost/math/distributions/chi_squared.hpp>
 
 // SeqAn3
 #include <seqan3/io/sam_file/all.hpp>
 #include <seqan3/core/debug_stream.hpp>
 
 // Class
 #include "IBPTree.hpp"
+#include "Stats.hpp"
 
 // define tags
 using seqan3::operator""_tag;
 
 namespace po = boost::program_options;
 namespace fs = boost::filesystem;
 namespace pt = boost::property_tree;
+namespace jp = boost::property_tree::json_parser;
 namespace ma = boost::math;
 
 class Analysis {
@@ -44,10 +47,12 @@ class Analysis {
         void normalize(); // normalize the frequencies to 1
 
         // write output files (of the analysis)
+        void writeStats();
         void writeInteractionsHeader(std::ofstream& fout);
-        void writeAllIntsHeader(std::ofstream& fout);
-        void addToAllIntsHeader(std::ofstream& fout, std::string key);
+        void writeAllIntsHeader(std::vector<int> condLastFlag, std::ofstream& fout);
         void writeAllInts();
+        void writeAllIntsCounts();
+        void writeAllIntsJGF();
 
         // other operations
         void addToFreqMap(std::pair<std::string,std::string> key, double value);
@@ -59,19 +64,27 @@ class Analysis {
         double calcGCS(std::vector<double>& complementarities);
         double calcGHS(std::vector<double>& hybenergies);
         double calcStat(dtp::IntKey key, int x);
+        double calcAdjusted(std::vector<double>& values);
 
     private:
         po::variables_map params;
         IBPTree features;
         std::map<std::pair<std::string,std::string>,double> freq; // strand, name
         std::string condition; // buffers the current condition
-        // maps for storing filters and suppreads
+        std::vector<std::string> conditions; // buffers all conditions
+
+        // maps for storing filters and suppreads (and other information)
         std::map<dtp::IntKey, std::vector<double>> suppreads;
         std::map<dtp::IntKey, std::vector<std::vector<double>>> complementarities;
         std::map<dtp::IntKey, std::vector<std::vector<double>>> hybenergies;
 
+        // Stats
+        std::shared_ptr<Stats> stats;
+
         int repcount; // counter for current replicate
         int readcount; // total number of reads
+        std::vector<int> repcountCond; // number of replicates per condition
+
 };
 
 #endif //RNANUE_ANALYSIS_HPP

include/DataTypes.hpp

@@ -75,9 +75,13 @@ namespace dtp {
         int alignedCount;
         int splitsCount;
         int multSplitsCount;
-        int nSurvivedCount;
+        int interactionsCount;
+        StatsFields() : readsCount(0), alignedCount(0), splitsCount(0), multSplitsCount(0), interactionsCount(0) {}
+        StatsFields(int readsCount, int alignedCount, int splitsCount, int multSplitsCount) :
+            readsCount(readsCount), alignedCount(alignedCount), splitsCount(splitsCount),
+            multSplitsCount(multSplitsCount), interactionsCount(0) {} // constructor for analysis class
     };
-    using StatsMap = std::map<std::string, StatsFields>;
+    using StatsMap = std::map<std::string, std::vector<StatsFields>>;
     using SpliceJunctions = std::map<std::string, std::vector<std::pair<size_t,size_t>>>;
 
     // Analysis

include/SplitReadCalling.hpp

@@ -111,13 +111,14 @@ class SplitReadCalling {
         void addComplementarityToSamRecord(SAMrecord &rec1, SAMrecord &rec2, TracebackResult &res);
         void addHybEnergyToSamRecord(SAMrecord &rec1, SAMrecord &rec2, double &hyb);
         void writeSAMrecordToBAM(auto& bamfile, std::vector<std::pair<SAMrecord, SAMrecord>>& records);
+        void writeStats();
 
 
     private:
         po::variables_map params;
         IBPTree features;
-        //Stats stats;
         std::shared_ptr<Stats> stats;
+        int replPerCond; // number of replicates per condition
         std::string condition; // stores the current condition
         std::deque<std::string> refIds; // stores the reference ids
         FilterScores filterScores;

include/Stats.hpp

@@ -18,6 +18,7 @@ namespace fs = boost::filesystem;
 class Stats {
     public:
         Stats();
+        Stats(std::string statsFile);
 
         // Move constructor / no needed because its unique
         Stats(Stats&& other) noexcept : stats(other.stats) {}
@@ -36,13 +37,16 @@ class Stats {
         Stats& operator=(const Stats&) = delete;
 
         // getter & setter
-        void setReadsCount(std::string condition, int increment);
-        void setAlignedCount(std::string condition, int increment);
-        void setSplitsCount(std::string condition, int increment);
-        void setMultSplitsCount(std::string condition, int increment);
+        void setReadsCount(std::string condition, int repl, int increment);
+        void setAlignedCount(std::string condition, int repl, int increment);
+        void setSplitsCount(std::string condition, int repl, int increment);
+        void setMultSplitsCount(std::string condition, int repl, int increment);
+        void setInteractionsCount(std::string condition, int repl, int increment);
+
+        void reserveStats(std::string condition, int repl); // creates new entry for replicate
 
         // write stats back to file
-        void writeStats(fs::path outdir);
+        void writeStats(fs::path outdir, std::string subcall);
 
     private:
         dtp::StatsMap stats;

include/Utility.hpp

@@ -5,6 +5,8 @@
 #include <iostream>
 #include <iomanip>
 #include <chrono>
+#include <random>
+
 
 // Boost
 #include <boost/filesystem.hpp>
@@ -42,10 +44,13 @@ namespace helper {
     bool withinRange(int a, int b, int range);
     std::string removeNonPrintable(const std::string str);
     std::string getTime(); // reports the current time
+
+    std::vector<int> lastOccFlag(std::vector<std::string>& vec);
 }
 
 namespace stats {
     double median(std::vector<double>& values);
+    double randNum(double min, double max);
 }
 
 // sequence input/output