Squashed commit of the following:

commit 3ec3f4ddde836f9c1112072a91536d8fc03df2e8 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Mon Mar 6 12:04:28 2023 +0100 Added line number to error message. commit 5f41e11788ce9ab09d0fd7e477b1bdeb0b16d952 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Mon Mar 6 11:36:10 2023 +0100 Fixed blastdb bug. commit e3725268db31c8c8ef652d61fda610defc8056b1 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Mon Mar 6 11:12:47 2023 +0100 Fix cmake error. commit b53a536c7e24ecba6778ef59ddf612aeb88c3c07 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Mon Mar 6 10:54:55 2023 +0100 Detect c++17. commit 8da058753e7894eb005ddbd0afa98782f1d0a1ce Author: Benjamin Buchfink <buchfink@gmail.com> Date: Mon Mar 6 10:33:27 2023 +0100 Fixed prepdb error. commit a2eb7c681cd2e34d404ec9c1821160d441c9e061 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 17:00:20 2023 +0100 Fixed error. commit 4a530f31297f53b3d2c9d7c5586e566cb2adfeab Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 16:48:26 2023 +0100 Fixed error. commit 6762d6c4f9ec277f8a8606f8dc1ac3069d95bbfa Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 16:07:07 2023 +0100 Fixed error. commit 2aa45bbbbacf33a24568a38ddb9ae5bcba39aee3 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 16:03:10 2023 +0100 Added mapping table to file. commit 44a8ffe6d2e08de4136e3c32265c646605285808 Merge: e6988560 2845fbc9 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 13:34:26 2023 +0100 Merge branch 'dev' of https://github.com/bbuchfink/diamond_dev into dev commit e6988560ed5d66a4d8d5e64c415963837d4957fe Author: Benjamin Buchfink <buchfink@gmail.com> Date: Fri Mar 3 13:34:22 2023 +0100 Print message to log. commit 2845fbc9e3f79dd7a7726fe79c19cccf7eac48fa Author: Dimi99 <73211787+Dimi99@users.noreply.github.com> Date: Fri Mar 3 13:33:43 2023 +0100 Dev (#11) * update remote * update remote * update remote * update remote * update remote * update remote * update remote * wfa libary * added the wfa2 library * added compile flags * change parameters * cleanup * bugfix * fix * removed debug print * repair merging errors * minor * adjusted preprocessor directives * changed size_t to Loc * removed std::make_unique * changed size_t to Loc * removed std::make_unique * added newline to the end of file * applied changes commit 2085a3a19acf923e32ef6e79f03b20ef2a29320f Author: Benjamin Buchfink <buchfink@gmail.com> Date: Thu Mar 2 16:54:20 2023 +0100 Added timers. commit a8277297aca52369970b2ac015dc7cbc3599b4e5 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Thu Mar 2 16:08:07 2023 +0100 Added cut command. commit cc564d9aa3ea309f6911d73a9de730b98feda8e0 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Thu Mar 2 15:36:16 2023 +0100 Added fle size message. commit e3fafe7fbaac267c3ab525628fada436b0be917b Author: Benjamin Buchfink <buchfink@gmail.com> Date: Thu Mar 2 14:59:18 2023 +0100 Added table building. commit 9570f9ad5ff48e1c3de32124c75065487621ffe3 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Thu Mar 2 10:40:22 2023 +0100 Added table building. commit c6038ab0b39f70936dbc0b2f07392f702481972a Author: Benjamin Buchfink <buchfink@gmail.com> Date: Wed Mar 1 17:00:40 2023 +0100 Added build helper. commit e0a5028e0777b89fc551315c69356dc0440cfc7f Author: Benjamin Buchfink <buchfink@gmail.com> Date: Wed Mar 1 15:34:32 2023 +0100 optimized tsv reading. commit 4db1c008ff05493364ebce7bcaaf9c9d87b0dc15 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Wed Mar 1 12:18:00 2023 +0100 Fixed bug. commit 18ae3ed01ea86b0623119b20e7c4307277be7736 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Wed Mar 1 12:05:58 2023 +0100 Optimized read_tsv. commit f0adbbd7c7d130726ef86f01fd5bf0f9ddf4dc24 Author: Dimi <dimitrios_K@gmx.de> Date: Wed Mar 1 09:32:23 2023 +0100 Revert "Merge branch 'dev' into dev" This reverts commit b0063bcc9947fa25015b5305065363333c324dfb, reversing changes made to 2f0537d067b668b2d2c4bcd64a4aa70e6dace604. commit b0063bcc9947fa25015b5305065363333c324dfb Merge: 2f0537d0 a74ab968 Author: Dimi99 <73211787+Dimi99@users.noreply.github.com> Date: Wed Mar 1 09:20:32 2023 +0100 Merge branch 'dev' into dev commit 2f0537d067b668b2d2c4bcd64a4aa70e6dace604 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Tue Feb 28 19:27:03 2023 +0100 Added read size option. commit 1bdef78d264061cd0416e9287f4a789587974351 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Tue Feb 28 18:55:57 2023 +0100 Added line counting. commit fc15dee3f2218997eaad2b3c024305e36a7177da Merge: 56950f13 dbd6bd2 Author: Benjamin Buchfink <buchfink@gmail.com> Date: Tue Feb 28 16:25:47 2023 +0100 Merge branch 'master' into dev
bbuchfink · Mar 6, 2023 · 40e440f · 40e440f
1 parent dbd6bd2
commit 40e440f
Show file tree

Hide file tree

Showing 316 changed files with 479,420 additions and 2,770 deletions.
diff --git a/CMakeLists.txt b/CMakeLists.txt
diff --git a/src/ChangeLog b/src/ChangeLog
@@ -1,3 +1,11 @@
+[2.1.5]
+- Disabled the use of frequency based seed masking when using the linear-time
+  search feature with respect to the targets.
+- Fixed a bug that caused a `Database file is not a BLAST database` error message
+  for the `prepdb` workflow.
+- Fixed a bug that caused a segmentation fault when using BLAST databases.
+- Added line numbers for error messages when reading taxonomy mapping files.
+
 [2.1.4]
 - Leading spaces are now trimmed and tabulator characters escaped as `\t`
   in sequence titles, and a warning message is produced.

diff --git a/src/align/align.cpp b/src/align/align.cpp
@@ -33,6 +33,7 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 #include "extend.h"
 #ifdef WITH_DNA
 #include "../dna/wfa2_test.h"
+#include "../dna/ksw2_extension.h"
 #endif
 #include "../util/algo/radix_sort.h"
 #include "target.h"
@@ -58,7 +59,7 @@ TextBuffer* pipeline_short(BlockId query, Search::Hit* begin, Search::Hit* end,
 
 }
 
-#ifndef EXTRA
+#if !defined(EXTRA) || defined(WITH_DNA)
 #define OLD
 #endif
 
@@ -253,7 +254,7 @@ bool align_worker(HitIterator* hit_it, ThreadPool::TaskSet* task_set, Search::Co
 
 			pair<vector<Extension::Match>, Extension::Stats> matches =
 #ifdef WITH_DNA
-				align_mode.mode == Align_mode::blastn ? WaveExtension::extend(*cfg, h->query) :
+           align_mode.mode == AlignMode::blastn ? Dna::extend(*cfg, cfg->query->seqs()[h->query]) :
 #endif
 				Extension::extend(h->query, h->begin, h->end, *cfg, stat, parallel ? DP::Flags::PARALLEL : DP::Flags::NONE);
 			TextBuffer* buf = cfg->blocked_processing ? Extension::generate_intermediate_output(matches.first, h->query, *cfg) : Extension::generate_output(matches.first, matches.second, h->query, stat, *cfg);
@@ -312,7 +313,7 @@ void align_queries(Consumer* output_file, Search::Config& cfg)
 		cfg.seed_hit_buf->load(std::min(mem_limit - res_size, config.trace_pt_fetch_size));
 
 		if (res_size + last_size > mem_limit)
-			message_stream << "Warning: resident size (" << (res_size + last_size) << ") exceeds memory limit." << std::endl;
+			log_stream << "Warning: resident size (" << (res_size + last_size) << ") exceeds memory limit." << std::endl;
 
 		timer.go("Sorting trace points");
 		ips4o::parallel::sort(hit_buf->begin(), hit_buf->end(), std::less<Search::Hit>(), config.threads_);
@@ -338,7 +339,7 @@ void align_queries(Consumer* output_file, Search::Config& cfg)
 		};
 #ifndef OLD
 		cfg.thread_pool.reset(new ThreadPool(task));
-		cfg.thread_pool->run(n_threads, !config.no_heartbeat);
+		cfg.thread_pool->run(n_threads, false);
 		cfg.thread_pool->join();
 #else
 		cfg.thread_pool.reset(new ThreadPool ());

diff --git a/src/basic/config.cpp b/src/basic/config.cpp
@@ -246,6 +246,8 @@ Config::Config(int argc, const char **argv, bool check_io, CommandLineParser& pa
 		.add_command("fetch-seq", "", FETCH_SEQ)
         .add_command("blastn", "Align DNA query sequences against a DNA reference database", blastn)
 		.add_command("length-sort", "", LENGTH_SORT)
+		.add_command("wc", "", WORD_COUNT)
+		.add_command("cut", "", CUT)
 #endif
 		;
 
@@ -334,7 +336,9 @@ Config::Config(int argc, const char **argv, bool check_io, CommandLineParser& pa
 \t5   = BLAST XML\n\
 \t6   = BLAST tabular\n\
 \t100 = DIAMOND alignment archive (DAA)\n\
-\t101 = SAM\n\n\
+\t101 = SAM\n\
+\t102 = Taxonomic classification\n\
+\t103 = PAF\n\n\
 \tValue 6 may be followed by a space-separated list of these keywords:\n\n\
 \tqseqid means Query Seq - id\n\
 \tqlen means Query sequence length\n\
@@ -401,7 +405,7 @@ Config::Config(int argc, const char **argv, bool check_io, CommandLineParser& pa
 
 	string algo_str;
 
-	auto& advanced = parser.add_group("Advanced options", { blastp, blastx, makeidx, CLUSTER_REASSIGN, regression_test, cluster, DEEPCLUST, LINCLUST });
+	auto& advanced = parser.add_group("Advanced options", { blastp, blastx, blastn, makeidx, CLUSTER_REASSIGN, regression_test, cluster, DEEPCLUST, LINCLUST });
 	advanced.add()
 		("algo", 0, "Seed search algorithm (0=double-indexed/1=query-indexed/ctg=contiguous-seed)", algo_str)
 		("bin", 0, "number of query bins for seed search", query_bins_)
@@ -445,6 +449,9 @@ Config::Config(int argc, const char **argv, bool check_io, CommandLineParser& pa
 		("log-evalue-scale", 0, "", log_evalue_scale, 1.0 / std::log(2.0))
 		("bootstrap", 0, "", bootstrap)
 		("mp-self", 0, "", mp_self)
+#ifdef EXTRA
+            ("zdrop", 'z', "zdrop for gapped dna alignment", zdrop, 40)
+#endif
 		("query-or-subject-cover", 0, "", query_or_target_cover);
 
 	auto& view_options = parser.add_group("View options", { view, blastp, blastx });
@@ -646,6 +653,7 @@ Config::Config(int argc, const char **argv, bool check_io, CommandLineParser& pa
 		("recluster_bd", 0, "", recluster_bd)
 		("pipeline-short", 0, "", pipeline_short)
 		("graph-algo", 0, "", graph_algo, string("gvc"))
+		("tsv-read-size", 0, "", tsv_read_size, int64_t(GIGABYTES))
 #ifndef KEEP_TARGET_ID
 		("kmer-ranking", 0, "Rank sequences based on kmer frequency in linear stage", kmer_ranking)
 #endif

diff --git a/src/basic/config.h b/src/basic/config.h
@@ -335,6 +335,8 @@ struct Config
 	bool pipeline_short;
 	string graph_algo;
 	bool linsearch;
+	int64_t tsv_read_size;
+    int zdrop;
 
     SequenceType dbtype;
 
@@ -350,7 +352,7 @@ struct Config
 		match_file_stat = 14, model_seqs = 15, opt = 16, mask = 17, fastq2fasta = 18, dbinfo = 19, test_extra = 20, test_io = 21, db_annot_stats = 22, read_sim = 23, info = 24, seed_stat = 25,
 		smith_waterman = 26, cluster = 27, translate = 28, filter_blasttab = 29, show_cbs = 30, simulate_seqs = 31, split = 32, upgma = 33, upgma_mc = 34, regression_test = 35,
 		reverse_seqs = 36, compute_medoids = 37, mutate = 38, rocid = 40, makeidx = 41, find_shapes, prep_db, composition, JOIN, HASH_SEQS, LIST_SEEDS, CLUSTER_REALIGN,
-		GREEDY_VERTEX_COVER, INDEX_FASTA, FETCH_SEQ, CLUSTER_REASSIGN, blastn, RECLUSTER, LENGTH_SORT, MERGE_DAA, DEEPCLUST, LINCLUST
+		GREEDY_VERTEX_COVER, INDEX_FASTA, FETCH_SEQ, CLUSTER_REASSIGN, blastn, RECLUSTER, LENGTH_SORT, MERGE_DAA, DEEPCLUST, LINCLUST, WORD_COUNT, CUT
 	};
 	unsigned	command;
 

diff --git a/src/basic/hssp.cpp b/src/basic/hssp.cpp
@@ -286,15 +286,22 @@ void Hsp::push_gap(Edit_operation op, int length, const Letter *subject)
 #endif
 }
 
-Hsp::Hsp(const IntermediateRecord &r, unsigned query_source_len, Loc qlen, Loc tlen, const OutputFormat* output_format) :
+Hsp::Hsp(const IntermediateRecord &r, unsigned query_source_len, Loc qlen, Loc tlen, const OutputFormat* output_format, const Stats::Blastn_Score *dna_score_builder):
 	backtraced(!IntermediateRecord::stats_mode(output_format->hsp_values) && output_format->hsp_values != HspValues::NONE),
 	score(r.score),
 	evalue(r.evalue),
 	bit_score(score_matrix.bitscore(r.score)),
 	corrected_bit_score(score_matrix.bitscore_corrected(r.score, qlen, tlen)),
 	transcript(r.transcript)
 {
-	subject_range.begin_ = r.subject_begin;
+    if(dna_score_builder == nullptr){
+        bit_score = score_matrix.bitscore(r.score);
+        corrected_bit_score = score_matrix.bitscore_corrected(r.score, qlen, tlen);
+    }
+    else
+        bit_score = dna_score_builder->blast_bit_Score(r.score);
+
+    subject_range.begin_ = r.subject_begin;
 	if (align_mode.mode == AlignMode::blastx) {
 		frame = r.frame(query_source_len, align_mode.mode);
 		set_translated_query_begin(r.query_begin, query_source_len);

diff --git a/src/basic/match.h b/src/basic/match.h
@@ -48,6 +48,7 @@ struct OutputFormat;
 
 namespace Stats {
 struct TargetMatrix;
+struct Blastn_Score;
 }
 
 struct Hsp
@@ -93,7 +94,7 @@ struct Hsp
 		matrix(nullptr)
 	{}
 
-	Hsp(const IntermediateRecord &r, unsigned query_source_len, Loc qlen, Loc tlen, const OutputFormat* output_format);
+    Hsp(const IntermediateRecord &r, unsigned query_source_len, Loc qlen, Loc tlen, const OutputFormat* output_format, const Stats::Blastn_Score *dna_score_builder = nullptr);
 	Hsp(const ApproxHsp& h, Loc qlen, Loc tlen);
 
 	struct Iterator

diff --git a/src/basic/packed_loc.h b/src/basic/packed_loc.h
@@ -3,7 +3,7 @@ DIAMOND protein aligner
 Copyright (C) 2013-2020 Max Planck Society for the Advancement of Science e.V.
                         Benjamin Buchfink
                         Eberhard Karls Universitaet Tuebingen
-						
+
 Code developed by Benjamin Buchfink <benjamin.buchfink@tue.mpg.de>
 
 This program is free software: you can redistribute it and/or modify
@@ -28,42 +28,42 @@ along with this program.  If not, see <http://www.gnu.org/licenses/>.
 
 struct packed_uint40_t
 {
-	uint8_t		high;
-	uint32_t	low;
-	packed_uint40_t():
-		high (),
-		low ()
-	{ }
-	packed_uint40_t(uint64_t v):
-		high ((uint8_t)(v>>32)),
-		low ((uint32_t)(v&0xfffffffflu))
-	{ }
-	packed_uint40_t& operator=(uint32_t x) {
-		high = 0;
-		low = x;
-		return *this;
-	}
-	operator uint64_t() const
-	{ return (uint64_t(high) << 32) | low; }
-	operator int64_t() const {
-		return (int64_t(high) << 32) | (int64_t)low;
-	}
-	operator uint32_t() const {
-		return low;
-	}
-	operator int32_t() const {
-		return low;
-	}
-	bool operator==(const packed_uint40_t& rhs) const {
-		return high == rhs.high && low == rhs.low;
-	}
-	bool operator!=(const packed_uint40_t& rhs) const {
-		return high != rhs.high && low != rhs.low;
-	}
-	bool operator<(const packed_uint40_t &rhs) const
-	{ return high < rhs.high || (high == rhs.high && low < rhs.low); }
-	friend uint64_t operator-(const packed_uint40_t &x, const packed_uint40_t &y)
-	{ return (uint64_t)(x) - (uint64_t)(y); }
+    uint8_t		high;
+    uint32_t	low;
+    packed_uint40_t():
+            high (),
+            low ()
+    { }
+    packed_uint40_t(uint64_t v):
+            high ((uint8_t)(v>>32)),
+            low ((uint32_t)(v&0xfffffffflu))
+    { }
+    packed_uint40_t& operator=(uint32_t x) {
+        high = 0;
+        low = x;
+        return *this;
+    }
+    operator uint64_t() const
+    { return (uint64_t(high) << 32) | low; }
+    operator int64_t() const {
+        return (int64_t(high) << 32) | (int64_t)low;
+    }
+    operator uint32_t() const {
+        return low;
+    }
+    operator int32_t() const {
+        return low;
+    }
+    bool operator==(const packed_uint40_t& rhs) const {
+        return high == rhs.high && low == rhs.low;
+    }
+    bool operator!=(const packed_uint40_t& rhs) const {
+        return high != rhs.high && low != rhs.low;
+    }
+    bool operator<(const packed_uint40_t &rhs) const
+    { return high < rhs.high || (high == rhs.high && low < rhs.low); }
+    friend uint64_t operator-(const packed_uint40_t &x, const packed_uint40_t &y)
+    { return (uint64_t)(x) - (uint64_t)(y); }
 } PACKED_ATTRIBUTE ;
 
 typedef packed_uint40_t PackedLoc;

diff --git a/src/basic/value.h b/src/basic/value.h
@@ -71,6 +71,9 @@ struct ValueTraits
 #define AMINO_ACID_ALPHABET "ARNDCQEGHILKMFPSTWYVBJZX*_"
 #define AMINO_ACID_COUNT (int(sizeof(AMINO_ACID_ALPHABET) - 1))
 
+#define NUCLEOTIDE_ALPHABET "ACGTN"
+#define NUCLEOTIDE_COUNT (sizeof(NUCLEOTIDE_ALPHABET) -1)
+
 constexpr Letter MASK_LETTER = 23;
 constexpr Letter STOP_LETTER = 24;
 constexpr Letter SUPER_HARD_MASK = 25;

diff --git a/src/cluster/helpers.cpp b/src/cluster/helpers.cpp
@@ -218,7 +218,7 @@ void init_thresholds() {
 }
 
 File* open_out_tsv() {
-	File* file = new File(Schema{ Type::STRING, Type::STRING }, config.output_file, Flags::WRITE | Flags::OVERWRITE);
+	File* file = new File(Schema{ Type::STRING, Type::STRING }, config.output_file, Flags::WRITE);
 	if (Blast_tab_format::header_format(::Config::cluster) == Header::SIMPLE)
 		file->write_record("centroid", "member");
 	return file;

diff --git a/src/data/blastdb/blastdb.cpp b/src/data/blastdb/blastdb.cpp
@@ -88,11 +88,13 @@ list<CRef<CSeq_id>>::const_iterator best_id(const list<CRef<CSeq_id>>& ids) {
 
 BlastDB::BlastDB(const std::string& file_name, Metadata metadata, Flags flags, const ValueTraits& value_traits) :
 	SequenceFile(Type::BLAST, Alphabet::NCBI, flags, FormatFlags::TITLES_LAZY | FormatFlags::SEEKABLE | FormatFlags::LENGTH_LOOKUP, value_traits),
-	file_name_(file_name),	
+	file_name_(file_name),
 	db_(new CSeqDBExpert(file_name, CSeqDB::eProtein)),
 	oid_(0),
 	long_seqids_(false),
-	flags_(flags)
+	flags_(flags),
+	sequence_count_(db_->GetNumOIDs()),
+	sparse_sequence_count_(db_->GetNumSeqs())
 {
 	if (flag_any(metadata, Metadata::TAXON_NODES | Metadata::TAXON_MAPPING | Metadata::TAXON_SCIENTIFIC_NAMES | Metadata::TAXON_RANKS))
 		throw std::runtime_error("Taxonomy features are not supported for the BLAST database format.");
@@ -242,12 +244,12 @@ std::vector<Letter> BlastDB::dict_seq(DictId dict_id, const size_t ref_block) co
 
 int64_t BlastDB::sequence_count() const
 {
-	return db_->GetNumOIDs();
+	return sequence_count_;
 }
 
 int64_t BlastDB::sparse_sequence_count() const
 {
-	return db_->GetNumSeqs();	
+	return sparse_sequence_count_;
 }
 
 size_t BlastDB::letters() const

diff --git a/src/data/blastdb/blastdb.h b/src/data/blastdb/blastdb.h
@@ -67,6 +67,7 @@ struct BlastDB : public SequenceFile {
 	int oid_;
 	const bool long_seqids_;
 	const Flags flags_;
+	int64_t sequence_count_, sparse_sequence_count_;
 	BitVector oid_filter_;
 
 	friend void load_blast_seqid();

diff --git a/src/data/enum_seeds.h b/src/data/enum_seeds.h
@@ -48,7 +48,10 @@ Search::SeedStats enum_seeds_minimizer(SequenceSet* seqs, F* f, unsigned begin,
 			continue;
 		seqs->convert_to_std_alph(i);
 		const Sequence seq = (*seqs)[i];
-		Reduction::reduce_seq(seq, buf);
+        if (align_mode.mode != AlignMode::blastn)
+            Reduction::reduce_seq(seq, buf);
+        else
+            buf = seq.copy();
 		for (size_t shape_id = cfg.shape_begin; shape_id < cfg.shape_end; ++shape_id) {
 			const Shape& sh = shapes[shape_id];
 			if (seq.length() < sh.length_) continue;
@@ -78,6 +81,7 @@ void enum_seeds_hashed(SequenceSet* seqs, F* f, unsigned begin, unsigned end, co
 			const Shape& sh = shapes[shape_id];
 			if (seq.length() < sh.length_) continue;
 			const uint64_t shape_mask = sh.long_mask();
+			//const __m128i shape_mask = sh.long_mask_sse_;
 			HashedSeedIterator<BITS> it(seq, sh);
 			Loc j = 0;
 			while (it.good()) {
@@ -103,7 +107,7 @@ void enum_seeds_contiguous(SequenceSet* seqs, F* f, unsigned begin, unsigned end
 		const Sequence seq = (*seqs)[i];
 		if (seq.length() < It::length()) continue;
 		It it(seq);
-		Loc j = 0;
+		size_t j = 0;
 		while (it.good()) {
 			if (it.get(key))
 				if (filter->contains(key, 0))

diff --git a/src/data/seed_array.h b/src/data/seed_array.h
@@ -49,18 +49,25 @@ struct SeedArray
 #else
 			value(pos)
 #endif
-		{}
-		SeedOffset key;
-
-		struct GetKey {
-			uint32_t operator()(const Entry& e) const {
-				return e.key;
-			}
-		};
+        {}
+        bool operator<(const Entry& entry)const{
+            return  this->key < entry.key;
+        }
+        bool operator==(const Entry& entry)const{
+            return this->key == entry.key && this->value == entry.value;
+        }
+        SeedOffset key;
+
+        struct GetKey {
+            uint32_t operator()(const Entry& e) const {
+                return e.key;
+            }
+        };
 
 		SeedLoc value;
 		using Key = decltype(key);
 		using Value = decltype(value);
+        using value_type =  Entry;
 	} PACKED_ATTRIBUTE;
 
 	template<typename _filter>

diff --git a/src/data/sequence_file.cpp b/src/data/sequence_file.cpp
@@ -842,6 +842,7 @@ void prep_db() {
 	else
 		FastaFile::prep_db(config.database);
 #else
+	else
 		throw runtime_error("Database file is not a BLAST database");
 #endif
 }

diff --git a/src/data/taxon_list.cpp b/src/data/taxon_list.cpp
@@ -64,10 +64,15 @@ static void load_mapping_file(ExternalSorter<pair<string, TaxId>>& sorter)
 	string accession, last;
 
 	while (!f.eof() && (f.getline(), !f.line.empty())) {
-		if (format == 0)
-			Util::String::Tokenizer(f.line, "\t") >> Util::String::Skip() >> accession >> taxid;
-		else
-			Util::String::Tokenizer(f.line, "\t") >> accession >> taxid;
+		try {
+			if (format == 0)
+				Util::String::Tokenizer(f.line, "\t") >> Util::String::Skip() >> accession >> taxid;
+			else
+				Util::String::Tokenizer(f.line, "\t") >> accession >> taxid;
+		}
+		catch (Util::String::TokenizerException&) {
+			throw std::runtime_error("Malformed input in line " + std::to_string(f.line_count));
+		}
 
 		if (accession.empty())
 			throw std::runtime_error("Empty accession field in line " + std::to_string(f.line_count));