Commit 6d8bb449 authored by Eric Coissac's avatar Eric Coissac

delete old version

git-svn-id: https://www.grenoble.prabi.fr/svn/LECASofts/ecoPrimers/trunk@179 60f365c0-8329-0410-b2a4-ec073aeeaa1d
parent f9410e5a
EXEC=ecoPrimer
PRIMER_SRC= ecoprimer.c
PRIMER_OBJ= $(patsubst %.c,%.o,$(PRIMER_SRC))
SRCS= $(PRIMER_SRC)
LIB= -lecoPCR -lapat -lKMRK -lz -lm
LIBFILE= libapat/libapat.a \
libecoPCR/libecoPCR.a \
libKMRK/libKMRK.a
include global.mk
all: $(EXEC)
########
#
# ecoPrimer compilation
#
########
# executable compilation and link
ecoPrimer: $(PRIMER_OBJ) $(LIBFILE)
$(CC) $(LDFLAGS) -o $@ $< $(LIBPATH) $(LIB)
########
#
# library compilation
#
########
libapat/libapat.a:
$(MAKE) -C libapat
libecoPCR/libecoPCR.a:
$(MAKE) -C libecoPCR
libKMRK/libKMRK.a:
$(MAKE) -C libKMRK
########
#
# project management
#
########
clean:
rm -f *.o
rm -f $(EXEC)
$(MAKE) -C libapat clean
$(MAKE) -C libecoPCR clean
$(MAKE) -C libKMRK clean
\ No newline at end of file
MACHINE=MAC_OS_X
LIBPATH= -Llibapat -LlibecoPCR -LlibKMRK
MAKEDEPEND = gcc -D$(MACHINE) -M $(CPPFLAGS) -o $*.d $<
CC=gcc
CFLAGS= -W -Wall -O2 -g
default: all
%.o: %.c
$(CC) -D$(MACHINE) $(CFLAGS) -c -o $@ $<
%.P : %.c
$(MAKEDEPEND)
@sed 's/\($*\)\.o[ :]*/\1.o $@ : /g' < $*.d > $@; \
rm -f $*.d; [ -s $@ ] || rm -f $@
include $(SRCS:.c=.P)
\ No newline at end of file
This diff is collapsed.
#ifndef _KMRK_H_
#define _KMRK_H_
/********************************************
********************************************
**
** Declaration of struct
**
********************************************
********************************************/
#include <stdint.h>
#include "repseek_types.h"
#include "KMRK_mask.h"
/**
* Structure used to manipulate simultanously
* the v and n vector
*
*/
typedef struct {
int32_t size; /**< size of vectors */
int32_t seqCount; /**< count of concatenated sequences */
int32_t complement; /**< if seqCount > 1 and complement !=0
* then second sequence is the inverted complement
* strand of first one */
int32_t symbmax;
int32_t* v; /**< sequence vector */
int32_t* n; /**< linked list vector */
int32_t limits[1]; /**< array of end limits of concatenated
* sequences in v (array size is seqCount) */
} vn_type;
/********************************************
********************************************
**
** Declaration of public macro
**
********************************************
********************************************/
// Return a pointer to a vector from a vn_type structure
#define GETVECTOR(x,vector) (((x)->vector) - 1)
#define IS_MARKED(x,i) ((x)[i] < 0)
#define MARK(x,i) ((x)[i]) = -ABS((x)[i])
#define UNMARK(x,i) ((x)[i]) = ABS((x)[i])
#define SET(x,i,v) ((x)[i]) = (v)
// set and mark in one operation
#define SETMARKED(x,i,v) ((x)[i]) = -(v)
//internal macro
#define GET(x,i) ABS((x)[i])
// get symbole at position i in vector x
#define SYMBOLE(x,i) ((IS_MARKED((x),(i))) ? (i): (GET(x,i)))
/**
* Macro used to declare a pointer to a quorum function.
*
* @param name name of the pointer
*
*/
#define KMRK_QUORUM_FUNC_PTR(name) int32_t (*name)(vn_type* x, \
int32_t pos, \
int32_t count, \
int32_t countSeq)
/**
* Macro used to declare a pointer to an initialisation function.
*
* @param name name of the pointer
* @param quorum name used for the quorum assiciated function
*
* @see KMRK_QUORUM_FUNC_PTR
*
*/
#define KMRK_INIT_FUNC_PTR(name,quorum) vn_type* (*name)(char *sequence, \
int32_t complement, \
int32_t count, \
int32_t countSeq, \
int32_t *k, \
KMRK_QUORUM_FUNC_PTR(quorum),\
masked_area_table_t *mask)
/********************************************
********************************************
**
** Declaration of public functions
**
********************************************
********************************************/
/**
* Initialise a vn_type record from one sequence to run KMRK algorithm
*
* @param sequence pointer to a C string containing the sequence
* @param complement != 0 means that seq one and two are the two strands of
* the same sequence.
* @param count parameter count passed to the quorun function
* @param countSeq parametter countSeq passed to the quorun function
* @param k length of the word represented by each symbole of v.
* k is an output parametter
* @param quorum pointer to a quorum function
*
* @return a pointer to vn_type structure correctly initialized
* to be used by KMRK_RunKMRK
*
* @see KMRK_HashOneSequence
*/
vn_type* KMRK_InitOneSequence(char *sequence,
int32_t complement,
int32_t count,
int32_t countSeq,
int32_t *k,
KMRK_QUORUM_FUNC_PTR(quorum),
masked_area_table_t *mask);
/**
* Initialise a vn_type record from one sequence to run KMRK algorithme.
* In more than KMRK_InitOneSequence, KMRK_HashOneSequence construct
* word of len k with an hash algorithm. k used is a function of
* sequence size and alphabet size. If calculed k is superior to lmax
* then k = lmax.
*
* @param sequence pointer to a C string containing the sequence
* @param complement != 0 means that seq one and two are the two strands of
* the same sequence.
* @param count parametter count passed to the quorun function
* @param countSeq parametter countSeq passed to the quorun function
* @param k maximum length of the created word (input)
* length of the word represented by each symbole
* of v (output).
* @param quorum pointer to a quorum function
*
* @return a pointer to vn_type structure correctly initialized
* to be used by KMRK_RunKMRK
*
* @see KMRK_InitOneSequence
*/
vn_type* KMRK_HashOneSequence(char *sequence,
int32_t complement,
int32_t count,
int32_t countSeq,
int32_t *k,
KMRK_QUORUM_FUNC_PTR(quorum),
masked_area_table_t *mask);
/**
* An example of quorum function testing than a factor is
* present a least two times. Because of definition of this
* quorum function count and countSeq parametter have no meanning
* in this instance of quorum function
*
* @param x a pointer to vn_type structure to check
* @param pos position in n of the beginning of the linked list to test
* @param count minimal number of occurence of factor
* @param countSeq minimal number of sequences concerned
*
* @return 1 if quorum is ok 0 otherwise.
*/
int32_t KMRK_CoupleQuorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
/**
* An example of quorum function testing than a factor is
* present a least two times in the direct strand of a sequence or
* at least one time in the direct strand and one time in the reverse
* strand. Because of definition of this
* quorum function count and countSeq parametter have no meanning
* in this instance of quorum function
*
* @param x a pointer to vn_type structure to check
* @param pos position in n of the beginning of the linked list to test
* @param count minimal number of occurence of factor
* @param countSeq minimal number of sequences concerned
*
* @return 1 if quorum is ok 0 otherwise.
*/
int32_t KMRK_DirInvQuorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
/**
* An example of quorum function testing than a factor is
* present a least one time in the direct strand and one time in the reverse
* strand. Because of definition of this
* quorum function count and countSeq parametter have no meanning
* in this instance of quorum function
*
* @param x a pointer to vn_type structure to check
* @param pos position in n of the beginning of the linked list to test
* @param count minimal number of occurence of factor
* @param countSeq minimal number of sequences concerned
*
* @return 1 if quorum is ok 0 otherwise.
*/
int32_t KMRK_InvQuorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
int32_t KMRK_Int12Quorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
int32_t KMRK_IntInv12Quorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
int32_t KMRK_IntDirInv12Quorum(vn_type* x,
int32_t pos,
int32_t count,
int32_t countSeq);
/**
* realize one cycle of KMR.
*
* @param x a pointer to vn_type created by an intialisation
* function or returned by this function.
* @param k step used to join two words
* @param count parametter count passed to the quorun function
* @param countSeq parametter countSeq passed to the quorun function
* @param KMRK_QUORUM_FUNC_PTR quorum pointer to a quorum function
*/
void KMRK_RunKMRK(vn_type *x,
int32_t k,
int32_t count,
int32_t countSeq,
KMRK_QUORUM_FUNC_PTR(quorum));
/**
* realises serveral run of KMR cycle to make from a sequence
* a vn_type structure describing sequences of factors of a precise size.
*
* @param sequence pointer to a C string containing the sequence
* @param size word size to construct
* @param count parametter count passed to the quorun function
* @param countSeq parametter countSeq passed to the quorun function
* @param quorum pointer to a quorum function
* @param init pointer to a initialisation function
*
* @return a vn_type pointer to a structure containing sequences of factors
*/
vn_type *KMRK_RunTo(char *sequence,
int32_t size,
int32_t complement,
int32_t count,
int32_t countSeq,
KMRK_QUORUM_FUNC_PTR(quorum),
KMRK_INIT_FUNC_PTR(init,quorum),
masked_area_table_t *mask);
/**
* free memory associated to a vn_type pointer
*
* @param x a pointer to vn_type structure
*/
void KMRK_FreeVN(vn_type *x);
int32_t KMRK_upperCoupleCount(vn_type *x);
int32_t KMRK_upperInvertedCount(vn_type* x,int32_t wordsize);
int32_t KMRK_upperInterCount(vn_type* x,int32_t seq1,int32_t seq2,int32_t wordsize);
void KMRK_markStart(vn_type* x);
#endif //_KMRK_H_
This diff is collapsed.
#ifndef KMRK_Seeds_h
#define KMRK_Seeds_h
/********************************************
********************************************
**
** Declaration of struct
**
********************************************
********************************************/
#include <stdint.h>
#include <stdio.h>
#include "KMRK.h"
#define KMRK_SORT_SEEDS_FUNC_PTR(name) int32_t (*name)(SmallSeed_type*, \
SmallSeed_type*)
#define KMRK_DELTA_SEEDS_FUNC_PTR(name) int32_t (*name)(SmallSeed_type*)
/********************************************
********************************************
**
** Declaration of public functions
**
********************************************
********************************************/
AllSeeds_type *KMRK_allocSeeds(AllSeeds_type *AllSeeds,
int32_t size,
int8_t opt_dir,
int8_t opt_inv);
void KMRK_SetMultipleLenDirSeeds(SmallSeed_type* seeds,
int32_t nseeds,
int32_t wordSize,
AllSeeds_type *PtrAllSeeds);
void KMRK_freeSeeds(AllSeeds_type *AllSeeds);
void KMRK_compactSeeds(AllSeeds_type *AllSeeds);
void KMRK_pushSeed(AllSeeds_type *AllSeeds,
int32_t pos1,
int32_t pos2,
int32_t length,
int8_t dir);
AllSeeds_type* KMRK_enumerateDirectCouple(AllSeeds_type* Seeds,
int32_t expected,
int32_t wordSize,
vn_type* stack,
int32_t seq);
AllSeeds_type* KMRK_enumerateInvertedCouple(AllSeeds_type* Seeds,
int32_t expected,
int32_t wordSize,
vn_type* stack);
AllSeeds_type* KMRK_enumerateInterCouple(AllSeeds_type* Seeds,
int32_t seq1,
int32_t seq2,
int32_t expected,
int32_t wordSize,
vn_type* stack);
AllSeeds_type* KMRK_enumerateInterInvertedCouple(AllSeeds_type* Seeds,
int32_t seq2,
int32_t expected,
int32_t wordSize,
vn_type* stack);
/**
* Compare two seeds and return an integer less than, equal to or greater
* than zero considering the relative order of the two seeds. This
* version take into account only pos1 and pos2 of seeds without taking
* account of the sequences or of the relative direction
*
* @param s1 pointer to seed one
* @param s2 pointer to seed two
*
* @return a integer less than, equal to or greater than zero
*/
int32_t KMRK_cmpSeedsPos(SmallSeed_type *s1, SmallSeed_type *s2);
int32_t KMRK_cmpDeltaSeedsPos(SmallSeed_type *s1, SmallSeed_type *s2);
int32_t KMRK_cmpDeltaInvSeedsPos(SmallSeed_type *s1, SmallSeed_type *s2);
void KMRK_sortSeeds(SmallSeed_type* seeds,
int32_t nseeds,
KMRK_SORT_SEEDS_FUNC_PTR(compare));
AllSeeds_type* KMRK_get_seeds(char **seq,
int32_t SimpleSeqLen,
int16_t Lmin,
int8_t opt_dir,
int8_t opt_inv,
int8_t opt_verbose,
masked_area_table_t *mask);
AllSeeds_type* KMRK_get_seeds_2seqs(char **seq1,
char **seq2,
int32_t size1,
int32_t size2,
int16_t Lmin,
int8_t opt_dir,
int8_t opt_inv,
int8_t opt_verbose,
masked_area_table_t *mask);
/**
* Order an array of seeds by pos1,pos2
*
* @param seeds pointer to an array of Seed_type object to sort
* @param count count of element in the array
*/
void KMRK_sortSeedsByPos(Seed_type* seeds, int32_t count);
#endif /* KMRK_Seeds_h */
/*
* KMRK_mask.c
* repseek
*
* Created by Eric Coissac on 04/12/04.
* Copyright 2004 __MyCompanyName__. All rights reserved.
*
*/
#include "KMRK_mask.h"
#include <stdio.h>
#include <stdlib.h>
#include "memory.h"
#define MASKED_AREA_TABLE_SIZE(seqcount) (sizeof(masked_area_table_t) + (sizeof(masked_area_list_t*) * ((seqcount)-1)))
#define MASKED_AREA_LIST_SIZE(areacount) (sizeof(masked_area_list_t) + (sizeof(masked_area_t) * ((areacount)-1)))
#define AREA_COUNT_INIT (1000)
static masked_area_table_t *new_masked_area_table(int32_t seqcount, int32_t areacount);
static masked_area_list_t *new_masked_area_list(int32_t areacount);
static masked_area_list_t *realloc_masked_area_list(masked_area_list_t *list,int32_t areacount);
static int32_t push_area(masked_area_table_t* table,int32_t sequence,int32_t begin,int32_t end);
static void sort_area_table(masked_area_table_t* table);
static int32_t compare_area(const masked_area_t* v1,const masked_area_t* v2);
static int32_t search_area(const masked_area_t* v1,const masked_area_t* v2);
static masked_area_list_t *strip_area_list(masked_area_list_t* list);
static void strip_area_table(masked_area_table_t* table);
static masked_area_list_t *new_masked_area_list(int32_t areacount)
{
masked_area_list_t *list;
list = MyCalloc(1,MASKED_AREA_LIST_SIZE(areacount),"Not enougth memory for mask table");
list->reserved=areacount;
return list;
}
static masked_area_list_t *realloc_masked_area_list(masked_area_list_t *list,int32_t areacount)
{
list = MyRealloc(list,
MASKED_AREA_LIST_SIZE(areacount),
MASKED_AREA_LIST_SIZE(list->reserved),
"Not enougth memory for mask table");
list->reserved=areacount;
return list;
}
static masked_area_table_t *new_masked_area_table(int32_t seqcount, int32_t areacount)
{
masked_area_table_t *table;
int32_t i;
table = MyCalloc(1,MASKED_AREA_TABLE_SIZE(seqcount),"Not enougth memory for mask table");
table->seqcount=seqcount;
for (i=0;i<seqcount;i++)
table->sequence[i]=new_masked_area_list(areacount);
return table;
}
static int32_t push_area(masked_area_table_t* table,int32_t sequence,int32_t begin,int32_t end)
{
masked_area_list_t * list;
if (sequence >= table->seqcount)
return -1;
list = table->sequence[sequence];
if (list->reserved == list->count)
{
list = realloc_masked_area_list(list,list->reserved*2);
table->sequence[sequence]=list;
}
list->area[list->count].begin=begin;
list->area[list->count].end=end;
list->count++;
table->total++;
return table->total;
}
static int32_t compare_area(const masked_area_t* v1,const masked_area_t* v2)
{
return v1->begin - v2->begin;
}
static void sort_area_table(masked_area_table_t* table)
{
int32_t i;
for (i=0; i<table->seqcount;i++)
{
qsort(table->sequence[i]->area,
table->sequence[i]->count,
sizeof(masked_area_t),
(int (*)(const void *, const void *))compare_area);
}
}
static masked_area_list_t *strip_area_list(masked_area_list_t* list)
{
int32_t i;
int32_t j;
int32_t count;
int32_t newcount;
count = list->count;
newcount=count;
for (i=1;i<count;i++)
{
/* fprintf(stderr,"\n%d->%d %d->%d ==>",list->area[i-1].begin,list->area[i-1].end,list->area[i].begin,list->area[i].end); */
if ((list->area[i].begin-1) <= list->area[i-1].end)
{
/* fprintf(stderr," joined"); */
list->area[i].begin=list->area[i-1].begin;
list->area[i-1].begin=-1;
newcount--;
}
}
list->count=newcount;
for (i=0,j=0;i<count;i++)
{
if (list->area[i].begin>=0)
{
if (i!=j)
list->area[j]=list->area[i];
j++;
}
}
return realloc_masked_area_list(list,newcount);
}
static void strip_area_table(masked_area_table_t* table)
{
int32_t seq;
int32_t oldcount;
masked_area_list_t* list;
sort_area_table(table);
for (seq=0; seq < table->seqcount;seq++)
{
list = table->sequence[seq];
oldcount = list->count;
table->sequence[seq]=strip_area_list(list);
table->total-=oldcount - table->sequence[seq]->count;
}
}
static int32_t search_area(const masked_area_t* v1,const masked_area_t* v2)
{
int32_t pos;
pos = v1->begin;
if (pos < v2->begin)
return -1;
if (pos > v2->end)
return 1;
return 0;
}
masked_area_table_t *KMRK_ReadMaskArea(char *areafile,int32_t seqcount,int32_t complement)
{
FILE* area;
char buffer[1000];
char *ok;
int32_t begin;
int32_t end;
int32_t sequence;
int32_t column;
int32_t linecount;
masked_area_table_t *table;
if (complement > 0)
seqcount++;
else
complement=0;
area = fopen(areafile,"r");
linecount=0;
table=new_masked_area_table(seqcount,AREA_COUNT_INIT);
do {
linecount++;
ok = fgets(buffer,999,area);
if (ok)
{
column = sscanf(buffer,"%d %d %d",&begin,&end,&sequence);
if (column > 1 && begin <= end)
{
begin--;
end--;
if (column==3)
sequence--;
else
sequence=0;
if