Functions to retrieve genome reference sequences from a binary FASTA file. More...

#include <stdio.h>
#include <string.h>
#include "binsearch.h"
#include "variantkey.h"

Macros
#define	ALLELE_MAXSIZE 256
	Maximum allele length. More...

#define	NORM_WRONGPOS (-2)
	Normalization: Invalid position. More...

#define	NORM_INVALID (-1)
	Normalization: Invalid reference. More...

#define	NORM_OK (0)
	Normalization: The reference allele perfectly match the genome reference. More...

#define	NORM_VALID (1)
	Normalization: The reference allele is inconsistent with the genome reference (i.e. when contains nucleotide letters other than A, C, G and T). More...

#define	NORM_SWAP (1 << 1)
	Normalization: The alleles have been swapped. More...

#define	NORM_FLIP (1 << 2)
	Normalization: The alleles nucleotides have been flipped (each nucleotide have been replaced with its complement). More...

#define	NORM_LEXT (1 << 3)
	Normalization: Alleles have been left extended. More...

#define	NORM_RTRIM (1 << 4)
	Normalization: Alleles have been right trimmed. More...

#define	NORM_LTRIM (1 << 5)
	Normalization: Alleles have been left trimmed. More...

Functions
static void	mmap_genoref_file (const char file, mmfile_t mf)

static int	aztoupper (int c)

static void	prepend_char (const uint8_t pre, char string, size_t size)

static char	get_genoref_seq (mmfile_t mf, uint8_t chrom, uint32_t pos)

static int	check_reference (mmfile_t mf, uint8_t chrom, uint32_t pos, const char *ref, size_t sizeref)

static void	flip_allele (char *allele, size_t size)

static void	swap_sizes (size_t first, size_t second)

static void	swap_alleles (char first, size_t sizefirst, char second, size_t sizesecond)

static int	normalize_variant (mmfile_t mf, uint8_t chrom, uint32_t pos, char ref, size_t sizeref, char alt, size_t *sizealt)

static uint64_t	normalized_variantkey (mmfile_t mf, const char chrom, size_t sizechrom, uint32_t pos, uint8_t posindex, char ref, size_t sizeref, char alt, size_t sizealt, int *ret)
	Returns a normalized 64 bit variant key based on CHROM, POS, REF, ALT. More...

Detailed Description

The functions provided here allows to retrieve genome reference sequences from a binary version of a genome reference FASTA file.

The input reference binary files can be generated from a FASTA file using the resources/tools/fastabin.sh script.

Macro Definition Documentation

#define ALLELE_MAXSIZE 256

#define NORM_FLIP (1 << 2)

#define NORM_INVALID (-1)

#define NORM_LEXT (1 << 3)

#define NORM_LTRIM (1 << 5)

#define NORM_OK (0)

#define NORM_RTRIM (1 << 4)

#define NORM_SWAP (1 << 1)

#define NORM_VALID (1)

#define NORM_WRONGPOS (-2)

Function Documentation

static int aztoupper ( int c )

inlinestatic

Returns the uppercase version of the input character. Note that this is safe to be used only with a-z characters. All characters above 'a' will be changed.

Parameters

c	Character to uppercase.

Returns: Uppercased character

static int check_reference	(	mmfile_t	mf,
		uint8_t	chrom,
		uint32_t	pos,
		const char *	ref,
		size_t	sizeref
	)

inlinestatic

Check if the reference allele matches the reference genome data.

Parameters

mf	Structure containing the memory mapped file.
chrom	Encoded Chromosome number (see encode_chrom).
pos	Position. The reference position, with the first base having position 0.
ref	Reference allele. String containing a sequence of nucleotide letters.
sizeref	Length of the ref string, excluding the terminating null byte.

Returns

Positive number in case of success, negative in case of error:

0 the reference allele match the reference genome;
1 the reference allele is inconsistent with the genome reference (i.e. when contains nucleotide letters other than A, C, G and T);
-1 the reference allele don't match the reference genome;
-2 the reference allele is longer than the genome reference sequence.

static void flip_allele	(	char *	allele,
		size_t	size
	)

inlinestatic

Flip the allele nucleotides (replaces each letter with its complement). The resulting string is always in uppercase. Support extended nucleotide letters.

Parameters

allele	Allele. String containing a sequence of nucleotide letters.
size	Length of the allele string.

static char get_genoref_seq	(	mmfile_t	mf,
		uint8_t	chrom,
		uint32_t	pos
	)

inlinestatic

Returns the genome reference nucleotide at the specified chromosome and position.

Parameters

mf	Structure containing the memory mapped file.
chrom	Encoded Chromosome number (see encode_chrom).
pos	Position. The reference position, with the first base having position 0.

Returns: The nucleotide letter or 0 in case of invalid position.

static void mmap_genoref_file	(	const char *	file,
		mmfile_t *	mf
	)

inlinestatic

Memory map the genoref binary file.

Parameters

file	Path to the file to map.
mf	Structure containing the memory mapped file.

Returns: Returns the memory-mapped file descriptors.

static int normalize_variant	(	mmfile_t	mf,
		uint8_t	chrom,
		uint32_t *	pos,
		char *	ref,
		size_t *	sizeref,
		char *	alt,
		size_t *	sizealt
	)

inlinestatic

Normalize a variant. Flip alleles if required and apply the normalization algorithm described at: https://genome.sph.umich.edu/wiki/Variant_Normalization

Parameters

mf	Structure containing the memory mapped file.
chrom	Chromosome encoded number.
pos	Position. The reference position, with the first base having position 0.
ref	Reference allele. String containing a sequence of nucleotide letters.
sizeref	Length of the ref string, excluding the terminating null byte.
alt	Alternate non-reference allele string.
sizealt	Length of the alt string, excluding the terminating null byte.

Returns

Positive bitmask number in case of success, negative number in case of error. When positive, each bit has a different meaning when set, has defined by the NORM_* defines:

bit 0 (NORM_VALID) : The reference allele is inconsistent with the genome reference (i.e. when contains nucleotide letters other than A, C, G and T).
bit 1 (NORM_SWAP) : The alleles have been swapped.
bit 2 (NORM_FLIP) : The alleles nucleotides have been flipped (each nucleotide have been replaced with its complement).
bit 3 (NORM_LEXT) : Alleles have been left extended.
bit 4 (NORM_RTRIM) : Alleles have been right trimmed.
bit 5 (NORM_LTRIM) : Alleles have been left trimmed.

static uint64_t normalized_variantkey	(	mmfile_t	mf,
		const char *	chrom,
		size_t	sizechrom,
		uint32_t *	pos,
		uint8_t	posindex,
		char *	ref,
		size_t *	sizeref,
		char *	alt,
		size_t *	sizealt,
		int *	ret
	)

inlinestatic

Parameters

mf	Structure containing the memory mapped binary fasta file.
chrom	Chromosome. An identifier from the reference genome, no white-space or leading zeros permitted.
sizechrom	Length of the chrom string, excluding the terminating null byte.
pos	Position. The reference position.
posindex	Position index: 0 for 0-based, 1 for 1-based.
ref	Reference allele. String containing a sequence of nucleotide letters. The value in the pos field refers to the position of the first nucleotide in the String. Characters must be A-Z, a-z or *.
sizeref	Length of the ref string, excluding the terminating null byte.
alt	Alternate non-reference allele string. Characters must be A-Z, a-z or *.
sizealt	Length of the alt string, excluding the terminating null byte.
ret	Normalization return value (see: normalize_variant).

Returns: Normalized VariantKey 64 bit code.

static void prepend_char	(	const uint8_t	pre,
		char *	string,
		size_t *	size
	)

inlinestatic

Prepend a character to a string.

Parameters

pre	Character to prepend.
string	String to modify.
size	Input string length.

Returns: void

static void swap_alleles	(	char *	first,
		size_t *	sizefirst,
		char *	second,
		size_t *	sizesecond
	)

inlinestatic

static void swap_sizes	(	size_t *	first,
		size_t *	second
	)

inlinestatic

Macros

Functions

Detailed Description

Macro Definition Documentation

Function Documentation