[med-svn] [Git][med-team/muscle][upstream] New upstream version 5.1.0
Lance Lin (@linqigang)
gitlab at salsa.debian.org
Sat Nov 5 12:21:22 GMT 2022
Lance Lin pushed to branch upstream at Debian Med / muscle
Commits:
9696cf25 by Lance Lin at 2022-11-05T19:11:32+07:00
New upstream version 5.1.0
- - - - -
16 changed files:
- .github/workflows/build_linux.yml
- README.md
- src/Makefile
- src/allocflat.cpp
- src/countsort.h
- src/ensemble.cpp
- src/gitver.bash
- src/gitver.bat
- src/help.h
- + src/help.txt
- src/myopts.h
- src/myutils.cpp
- src/myutils.h
- src/testlog.cpp
- src/timing.h
- src/usage.h
Changes:
=====================================
.github/workflows/build_linux.yml
=====================================
@@ -21,7 +21,7 @@ jobs:
cd $GITHUB_WORKSPACE
ls -lh
cd src
- make
+ make LDFLAGS2=-static
- name: Upload binary artifact
uses: actions/upload-artifact at v2
with:
=====================================
README.md
=====================================
@@ -17,6 +17,10 @@ https://github.com/rcedgar/muscle/releases
[Muscle v5 home page](https://drive5.com/muscle5)
[Manual](https://drive5.com/muscle5/manual)
+### Building MUSCLE from source
+
+[https://github.com/rcedgar/muscle/wiki/Building-MUSCLE](https://github.com/rcedgar/muscle/wiki/Building-MUSCLE)
+
### Reference
R.C. Edgar (2021) "MUSCLE v5 enables improved estimates of phylogenetic tree confidence by ensemble bootstrapping"
=====================================
src/Makefile
=====================================
@@ -26,10 +26,7 @@ endif
CXXFLAGS := $(CXXFLAGS) -O3 -fopenmp -ffast-math
-LDFLAGS := $(LDFLAGS) -O3 -fopenmp -pthread -lpthread
-ifeq ($(OS),Linux)
- LDFLAGS += -static
-endif
+LDFLAGS := $(LDFLAGS) -O3 -fopenmp -pthread -lpthread ${LDFLAGS2}
HDRS := $(shell echo *.h)
OBJS := $(shell echo *.cpp | sed "-es/^/$(OS)\//" | sed "-es/ / $(OS)\//g" | sed "-es/\.cpp/.o/g")
=====================================
src/allocflat.cpp
=====================================
@@ -3,6 +3,8 @@
uint64 GetFBSize(uint LX, uint LY)
{
uint64 Size64 = uint64(LX + 1)*uint64(LY + 1)*HMMSTATE_COUNT;
+ if (double(Size64) > 4e9)
+ Die("Memory object too large due to sequence lengths %u, %u", LX, LY);
uint Size = uint(Size64);
asserta(Size == uint(Size64));
return Size;
=====================================
src/countsort.h
=====================================
@@ -20,7 +20,7 @@ public:
CountSortMem()
{
m_MaxValueCount = 0;
- zero(m_Vecs, NVEC);
+ memset_zero(m_Vecs, NVEC);
}
void Free()
=====================================
src/ensemble.cpp
=====================================
@@ -41,7 +41,7 @@ void Ensemble::SortMSA(MSA &M)
M.GetLabelToSeqIndex(Labels2, LabelToSeqIndex2);
char **szSeqsSorted = myalloc(char *, SeqCount);
- zero(szSeqsSorted, SeqCount);
+ memset_zero(szSeqsSorted, SeqCount);
for (uint SeqIndex = 0; SeqIndex < SeqCount; ++SeqIndex)
{
const string &Label = Labels2[SeqIndex];
=====================================
src/gitver.bash
=====================================
@@ -1,5 +1,13 @@
#!/bin/bash
+if [ ! -d ../.git ] ; then
+ if [ ! -f gitver.txt ] ; then
+ echo "0" > gitver.txt
+ fi
+ echo "Repo not found, git hash set to zero"
+ exit 0
+fi
+
PATH=$PATH:/usr/bin
git describe --abbrev=6 --dirty --long --always \
=====================================
src/gitver.bat
=====================================
@@ -3,7 +3,7 @@
if exist gitver.txt (
echo gitver.txt found
) else (
- echo 000 > gitver.txt
+ echo "-" > gitver.txt
)
if exist c:\cygwin64\bin\bash.exe (
=====================================
src/help.h
=====================================
@@ -1 +1,66 @@
-"This is help.txt\n"
+"Align FASTA input, write aligned FASTA (AFA) output:\n"
+" muscle -align input.fa -output aln.afa\n"
+"\n"
+"Align large input using Super5 algorithm if -align is too expensive,\n"
+"typically needed with more than a few hundred sequences:\n"
+" muscle -super5 input.fa -output aln.afa\n"
+"\n"
+"Single replicate alignment:\n"
+" muscle -align input.fa -perm PERM -perturb SEED -output aln.afa\n"
+" muscle -super5 input.fa -perm PERM -perturb SEED -output aln.afa\n"
+" PERM is guide tree permutation none, abc, acb, bca (default none).\n"
+" SEED is perturbation seed 0, 1, 2... (default 0 = don't perturb).\n"
+"\n"
+"Ensemble of replicate alignments, output in Ensemble FASTA (EFA) format,\n"
+"EFA has one aligned FASTA for each replicate with header line \"<PERM.SEED\":\n"
+" muscle -align input.fa -stratified -output stratified_ensemble.efa\n"
+" muscle -align input.fa -diversified -output diversified_ensemble.afa\n"
+"\n"
+" -replicates N\n"
+" Number of replicates, defaults 4, 100, 100 for stratified,\n"
+" diversified, resampled. With -stratified there is one\n"
+" replicate per guide tree permutation, total is 4 x N.\n"
+"\n"
+"Generate resampled ensemble from existing ensemble by sampling columns\n"
+"with replacement:\n"
+" muscle -resample ensemble.efa -output resampled.efa\n"
+"\n"
+" -max_gap_fract F\n"
+" Maximum fraction of gaps in a column (F=0..1, default 0.5).\n"
+"\n"
+" -minconf CC\n"
+" Minimum column confidence (CC=0..1, default 0.5).\n"
+"\n"
+"If ensemble output filename has @, then one FASTA file is generated\n"
+"for each replicate where @ is replaced by perm.s, otherwise all replicates\n"
+"are written to one EFA file.\n"
+"\n"
+"Calculate disperson of an ensemble:\n"
+" muscle -disperse ensemble.efa\n"
+"\n"
+"Extract replicate with highest total CC (diversified input recommended):\n"
+" muscle -maxcc ensemble.efa -output maxcc.afa\n"
+"\n"
+"Extract aligned FASTA files from EFA file:\n"
+" muscle -efa_explode ensemble.efa\n"
+"\n"
+"Convert FASTA to EFA, input has one filename per line:\n"
+" muscle -fa2efa filenames.txt -output ensemble.efa\n"
+"\n"
+"Update ensemble by adding two sequences of digits to each replicate, digits\n"
+"are column confidence (CC) values, e.g. \"73\" means CC=0.73, \"++\" is CC=1.0:\n"
+" muscle -addconfseq ensemble.efa -output ensemble_cc.efa\n"
+"\n"
+"Calculate letter confidence (LC) values, -ref specifies the alignment to\n"
+"compare against the ensemble (e.g. from -maxcc), output is in aligned\n"
+"FASTA format with LC values 0, 1 ... 9 instead of letters:\n"
+" muscle -letterconf ensemble.efa -ref aln.afa -output letterconf.afa\n"
+"\n"
+" -html aln.html\n"
+" Alignment colored by LC in HTML format.\n"
+"\n"
+" -jalview aln.features\n"
+" Jalview feature file with LC values and colors.\n"
+"\n"
+"More documentation at:\n"
+" https://drive5.com/muscle\n"
=====================================
src/help.txt
=====================================
@@ -0,0 +1,66 @@
+Align FASTA input, write aligned FASTA (AFA) output:
+ muscle -align input.fa -output aln.afa
+
+Align large input using Super5 algorithm if -align is too expensive,
+typically needed with more than a few hundred sequences:
+ muscle -super5 input.fa -output aln.afa
+
+Single replicate alignment:
+ muscle -align input.fa -perm PERM -perturb SEED -output aln.afa
+ muscle -super5 input.fa -perm PERM -perturb SEED -output aln.afa
+ PERM is guide tree permutation none, abc, acb, bca (default none).
+ SEED is perturbation seed 0, 1, 2... (default 0 = don't perturb).
+
+Ensemble of replicate alignments, output in Ensemble FASTA (EFA) format,
+EFA has one aligned FASTA for each replicate with header line "<PERM.SEED":
+ muscle -align input.fa -stratified -output stratified_ensemble.efa
+ muscle -align input.fa -diversified -output diversified_ensemble.afa
+
+ -replicates N
+ Number of replicates, defaults 4, 100, 100 for stratified,
+ diversified, resampled. With -stratified there is one
+ replicate per guide tree permutation, total is 4 x N.
+
+Generate resampled ensemble from existing ensemble by sampling columns
+with replacement:
+ muscle -resample ensemble.efa -output resampled.efa
+
+ -max_gap_fract F
+ Maximum fraction of gaps in a column (F=0..1, default 0.5).
+
+ -minconf CC
+ Minimum column confidence (CC=0..1, default 0.5).
+
+If ensemble output filename has @, then one FASTA file is generated
+for each replicate where @ is replaced by perm.s, otherwise all replicates
+are written to one EFA file.
+
+Calculate disperson of an ensemble:
+ muscle -disperse ensemble.efa
+
+Extract replicate with highest total CC (diversified input recommended):
+ muscle -maxcc ensemble.efa -output maxcc.afa
+
+Extract aligned FASTA files from EFA file:
+ muscle -efa_explode ensemble.efa
+
+Convert FASTA to EFA, input has one filename per line:
+ muscle -fa2efa filenames.txt -output ensemble.efa
+
+Update ensemble by adding two sequences of digits to each replicate, digits
+are column confidence (CC) values, e.g. "73" means CC=0.73, "++" is CC=1.0:
+ muscle -addconfseq ensemble.efa -output ensemble_cc.efa
+
+Calculate letter confidence (LC) values, -ref specifies the alignment to
+compare against the ensemble (e.g. from -maxcc), output is in aligned
+FASTA format with LC values 0, 1 ... 9 instead of letters:
+ muscle -letterconf ensemble.efa -ref aln.afa -output letterconf.afa
+
+ -html aln.html
+ Alignment colored by LC in HTML format.
+
+ -jalview aln.features
+ Jalview feature file with LC values and colors.
+
+More documentation at:
+ https://drive5.com/muscle
=====================================
src/myopts.h
=====================================
@@ -1,5 +1,5 @@
#ifndef MY_VERSION
-#define MY_VERSION "5.1"
+#define MY_VERSION "5.2"
#endif
#define PROGRAM_NAME "muscle"
=====================================
src/myutils.cpp
=====================================
@@ -807,7 +807,7 @@ static char *GetThreadStr()
{
unsigned NewThreadStrCount = ThreadIndex + 4;
char **NewThreadStrs = myalloc(char *, NewThreadStrCount);
- zero(NewThreadStrs, NewThreadStrCount);
+ memset_zero(NewThreadStrs, NewThreadStrCount);
if (g_ThreadStrCount > 0)
memcpy(NewThreadStrs, g_ThreadStrs, g_ThreadStrCount*sizeof(char *));
g_ThreadStrs = NewThreadStrs;
@@ -898,6 +898,8 @@ void Log(const char *Format, ...)
void Die_(const char *Format, ...)
{
+ va_list ArgList;
+ va_start(ArgList, Format);
#pragma omp critical
{
static bool InDie = false;
@@ -908,10 +910,7 @@ void Die_(const char *Format, ...)
if (g_fLog != 0)
setbuf(g_fLog, 0);
- va_list ArgList;
- va_start(ArgList, Format);
myvstrprintf(Msg, Format, ArgList);
- va_end(ArgList);
fprintf(stderr, "\n\n");
Log("\n");
@@ -944,6 +943,7 @@ void Die_(const char *Format, ...)
exit(1);
}
+ va_end(ArgList);
}
void Warning_(const char *Format, ...)
@@ -2076,7 +2076,8 @@ void GetVersionString(string &s)
"T"
#endif
;
-
+ if (GIT_VER == 0)
+ GIT_VER = "-";
Ps(s, "%s %s.%s%s [%s]", PROGRAM_NAME, MY_VERSION, GetPlatform(), Flags, GIT_VER);
}
@@ -2398,15 +2399,8 @@ void LogAllocSummary()
void *mymalloc64(unsigned BytesPerObject, uint64 N)
{
uint64 Bytes = N*BytesPerObject;
- //byte *p = 0;
- //try
- // {
- // p = new byte[Bytes];
- // }
- //catch (...)
- // {
- // Die("myalloc64(%u, %.3g) failed", BytesPerObject, double(N));
- // }
+ if (Bytes >= UINT32_MAX - 1024)
+ Die("Memory object >4Gb, probably due to long seqences");
byte *p = (byte *) malloc(Bytes);
if (p == 0)
Die("myalloc64(%u, %.3g) failed", BytesPerObject, double(N));
=====================================
src/myutils.h
=====================================
@@ -358,7 +358,6 @@ inline bool feq(double x, double y)
#define asserteq(x, y) assert(feq(x, y))
#define assertaeq(x, y) asserta(feq(x, y))
-#define zero(a, n) memset((a), 0, (n)*sizeof(a[0]))
#define memset_zero(a, n) memset((a), 0, (n)*sizeof(a[0]))
void ResetRand(unsigned Seed);
=====================================
src/testlog.cpp
=====================================
@@ -1,6 +1,10 @@
#include "muscle.h"
#include "timing.h"
+void cmd_testlog() {}
+
+#if 0
+
inline float HACK(float x)
{
assert(x >= 0.00f);
@@ -151,3 +155,4 @@ Next:
ProgressLog("LOG_ADD %.3g sum_log_prob %.3g lookup %.3g\n",
double(t2 - t1), double(t4 - t3), double(t6 - t5));
}
+#endif // 0
=====================================
src/timing.h
=====================================
@@ -1,6 +1,8 @@
#ifndef getticks_h
#define getticks_h
+#if 0
+
// ~3 x 10^9 ticks/sec
#ifdef _MSC_VER
@@ -31,4 +33,6 @@ __inline__ uint64_t GetClockTicks()
#error "getticks_h, unknown compiler"
#endif
+#endif
+
#endif // getticks_h
=====================================
src/usage.h
=====================================
@@ -49,7 +49,7 @@
"\n"
"Update ensemble by adding two sequences of digits to each replicate, digits\n"
"are column confidence (CC) values, e.g. \"73\" means CC=0.73, \"++\" is CC=1.0:\n"
-" muscle -addconfseqs ensemble.efa -output ensemble_cc.efa\n"
+" muscle -addconfseq ensemble.efa -output ensemble_cc.efa\n"
"\n"
"Calculate letter confidence (LC) values, -ref specifies the alignment to\n"
"compare against the ensemble (e.g. from -maxcc), output is in aligned\n"
View it on GitLab: https://salsa.debian.org/med-team/muscle/-/commit/9696cf2541762a27d1745699a24346f2933cd08c
--
View it on GitLab: https://salsa.debian.org/med-team/muscle/-/commit/9696cf2541762a27d1745699a24346f2933cd08c
You're receiving this email because of your account on salsa.debian.org.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://alioth-lists.debian.net/pipermail/debian-med-commit/attachments/20221105/8e861650/attachment-0001.htm>
More information about the debian-med-commit
mailing list