include/encoding.h File Reference

This graph shows which files directly or indirectly include this file:

Go to the source code of this file.

Data Structures
struct	wildcard
struct	wildcardEdit
Defines
#define	encoding_nucleotide   1
#define	encoding_protein   0
#define	encoding_aaStartWildcards   24
#define	encoding_extractBase(byte, bytePosition)   ((byte >> (6 - (bytePosition * 2))) & 0x3)
Functions
void	encoding_initialize (unsigned char alphabetType)
unsigned char	encoding_getComplement (unsigned char code)
unsigned char	encoding_getCode (unsigned char letter)
unsigned char	encoding_getLetter (unsigned char code)
unsigned char	encoding_determineAlphabetType (char *sequence, uint4 sequenceSize)
unsigned char	encoding_randomEncodedLetter (unsigned char code)
void	encoding_insertWilds (unsigned char *subject, unsigned char *edits, unsigned char *endEdits)
unsigned char *	encoding_byteUnpack (unsigned char *bytePackedSequence, int4 sequenceLength)
unsigned char	encoding_bytePack (unsigned char *sequence)
unsigned char	encoding_bytePackRemaining (unsigned char *sequence, int4 numLetters)
unsigned char	encoding_bytePackBeginning (unsigned char *sequence, int4 numLetters)
int4	encoding_replaceWildcards (struct memSingleBlock *wildcardEdits, unsigned char *sequence, int4 sequenceSize)
int4	encoding_bytePackSequence (unsigned char *sequence, int4 sequenceSize)
void	encoding_printLetters (unsigned char code, int4 numLetters)
void	encoding_encodeSequence (char *sequence, int4 sequenceSize, uint4 alphabetType)
void	encoding_free ()
unsigned char *	encoding_byteUnpackRegion (unsigned char *subject, unsigned char *bytePackedSequence, int4 sequenceLength)
Variables
packedByte *	encoding_packedByteLookup
unsigned char *	encoding_codesArray
unsigned char *	encoding_lettersArray
unsigned char	encoding_unknownCode
unsigned char	encoding_sentinalCode
unsigned char	encoding_numLetters
unsigned char	encoding_numRegularLetters
unsigned char	encoding_numCodes
unsigned char	encoding_alphabetType
wildcard *	encoding_wildcards
char *	encoding_alphabetTypes [2]

---

Define Documentation

#define encoding_aaStartWildcards   24

Definition at line 32 of file encoding.h.

Referenced by cluster_addChild(), cluster_calculateSavings(), cluster_score(), readdb_getChildren(), scoreMatrix_load(), wildcards_outputWildcards(), and wildcards_readWildcards().

#define encoding_extractBase	(	byte,
		bytePosition		)	((byte >> (6 - (bytePosition * 2))) & 0x3)

Definition at line 57 of file encoding.h.

Referenced by nuGappedScoring_dpAfterSeed(), nuGappedScoring_dpBeforeSeed(), and nuGappedScoring_score().

#define encoding_nucleotide   1

Definition at line 30 of file encoding.h.

Referenced by alignments_createNew(), blast_search(), determineDbAlphabetType(), dust_dustSequence(), encoding_determineAlphabetType(), encoding_free(), encoding_initialize(), gappedScoring_score(), hitMatrix_initialize(), hitMatrix_reinitialize(), main(), parameters_free(), parameters_loadDefaults(), print_constructAlignment(), PSSMatrix_create(), PSSMatrix_free(), statistics_calculateUngappedKarlinParameters(), unpack_free(), wordLookupDFA_build(), and writedb_addSequence().

#define encoding_protein   0

Definition at line 31 of file encoding.h.

Referenced by alignments_findGoodAlignments(), blast_search(), determineDbAlphabetType(), encoding_determineAlphabetType(), encoding_initialize(), encoding_randomEncodedLetter(), hitMatrix_free(), main(), print_constructAlignment(), print_gappedExtension(), print_XMLheader(), readdb_nextVolume(), readdb_open(), statistics_initialize(), ungappedExtension_findSeed(), unpack_loadSubject(), unpack_unpackSubject(), wordLookupDFA_print(), writedb_close(), and writedb_initialize().

---

Function Documentation

unsigned char encoding_bytePack

(

unsigned char *

sequence

)

[inline]

Definition at line 391 of file encoding.c.

Referenced by encoding_bytePackSequence(), nucleotideLookup_build(), and PSSMatrix_create().

{
        return (*sequence << 6) | (*(sequence + 1) << 4) | (*(sequence + 2) << 2) | *(sequence + 3);
}

Here is the caller graph for this function:

unsigned char encoding_bytePackBeginning	(	unsigned char *	sequence,
		int4	numLetters
	)			[inline]

Definition at line 418 of file encoding.c.

Referenced by PSSMatrix_create().

{
        if (numLetters == 1)
    {
                return *sequence;
        }
    else if (numLetters == 2)
    {
        return *(sequence + 1) | (*sequence << 2);
        }
    else if (numLetters == 3)
    {
        return *(sequence + 2) | (*(sequence + 1) << 2) | (*sequence << 4);
        }
    else // numLetters = 4
    {
                return *(sequence + 3) | (*(sequence + 2) << 2) | (*(sequence + 1) << 4) | (*sequence << 6);
    }
}

Here is the caller graph for this function:

unsigned char encoding_bytePackRemaining	(	unsigned char *	sequence,
		int4	numLetters
	)			[inline]

Definition at line 397 of file encoding.c.

Referenced by encoding_bytePackSequence(), and PSSMatrix_create().

{
        if (numLetters == 1)
    {
                return (*sequence << 6);
        }
    else if (numLetters == 2)
    {
                return (*(sequence + 1) << 4) | (*sequence << 6);
        }
    else if (numLetters == 3)
    {
                return (*(sequence + 2) << 2) | (*(sequence + 1) << 4) | (*sequence << 6);
        }
    else // numLetters = 4
    {
                return *(sequence + 3) | (*(sequence + 2) << 2) | (*(sequence + 1) << 4) | (*sequence << 6);
    }
}

Here is the caller graph for this function:

int4 encoding_bytePackSequence	(	unsigned char *	sequence,
		int4	sequenceSize
	)

Definition at line 469 of file encoding.c.

References encoding_bytePack(), encoding_bytePackRemaining(), and int4.

Referenced by writedb_addSequence().

{
        int4 count = 0;

    // For each group of four letters
    while (count < sequenceSize / 4)
    {
                sequence[count] = encoding_bytePack(sequence + count * 4);
        count++;
    }

    // Pack any remaining 1 to 3 letters
    if (sequenceSize % 4)
    {
                sequence[count] = encoding_bytePackRemaining(sequence + count * 4, sequenceSize % 4);
        count++;
    }

    return count;
}

Here is the call graph for this function:

Here is the caller graph for this function:

unsigned char* encoding_byteUnpack	(	unsigned char *	bytePackedSequence,
		int4	sequenceLength
	)

Definition at line 375 of file encoding.c.

References encoding_byteUnpackRegion(), and global_malloc().

Referenced by main(), and search_nucleotideSsearch().

{
        unsigned char* subject;

    subject = global_malloc(sizeof(char) * sequenceLength);
        encoding_byteUnpackRegion(subject, bytePackedSequence, sequenceLength);

    return subject;
}

Here is the call graph for this function:

Here is the caller graph for this function:

unsigned char* encoding_byteUnpackRegion	(	unsigned char *	subject,
		unsigned char *	bytePackedSequence,
		int4	sequenceLength
	)

Definition at line 321 of file encoding.c.

References packedByte::codes, encoding_packedByteLookup, and int4.

Referenced by encoding_byteUnpack(), unpack_extendRegionEnd(), unpack_extendRegionStart(), and unpack_unpackSubject().

{
        unsigned char *subjectPosition;
    int4 packedCount = 0;
    int4 numPackedBytes, numRemaining;

    subjectPosition = subject;

    // Calculate packed length and remainder
    numPackedBytes = sequenceLength / 4;
    numRemaining = sequenceLength % 4;

    // For each packed byte
        while (packedCount < numPackedBytes)
    {
        // Copy the value to the subject sequence
                memcpy(subjectPosition, encoding_packedByteLookup[bytePackedSequence[packedCount]].codes,
               sizeof(char) * 4);

        subjectPosition+=4;
        packedCount++;
    }

    // Unpack last 1-3 letters
    if (numRemaining)
    {
                if (numRemaining == 1)
        {
                *subjectPosition = (bytePackedSequence[packedCount] >> 6);
            packedCount++;
        }
        else if (numRemaining == 2)
        {
                *subjectPosition = (bytePackedSequence[packedCount] >> 6);
            subjectPosition++;
                *subjectPosition = (bytePackedSequence[packedCount] >> 4) & 0x3;
            packedCount++;
        }
        else if (numRemaining == 3)
        {
                *subjectPosition = (bytePackedSequence[packedCount] >> 6);
            subjectPosition++;
                *subjectPosition = (bytePackedSequence[packedCount] >> 4) & 0x3;
            subjectPosition++;
                *subjectPosition = (bytePackedSequence[packedCount] >> 2) & 0x3;
            packedCount++;
        }
    }

    return subject;
}

Here is the caller graph for this function:

unsigned char encoding_determineAlphabetType	(	char *	sequence,
		uint4	sequenceSize
	)

Definition at line 230 of file encoding.c.

References encoding_nucleotide, encoding_nuLettersArray, encoding_nuNumLetters, encoding_nuNumRegularLetters, encoding_protein, and int4.

Referenced by determineDbAlphabetType(), and main().

{
        int4 letterCount, alphabetCount, matches = 0, wildchars = 0;
    char letter;

    // Go through the sequence and check it for non-nucleic characters
    letterCount = 0;
    while (letterCount < sequenceSize)
    {
        letter = sequence[letterCount];

        // For each regular nucleic letter
        alphabetCount = 0;
        while (alphabetCount <= encoding_nuNumRegularLetters)
        {
            // Check for a match
            if (toupper(letter) == encoding_nuLettersArray[alphabetCount])
                matches++;

            alphabetCount++;
        }

        // For each wildcard character
        while (alphabetCount < encoding_nuNumLetters)
        {
            // Check for a match
            if (toupper(letter) == encoding_nuLettersArray[alphabetCount])
                wildchars++;

            alphabetCount++;
        }

        letterCount++;
    }

//    printf("[%d,%d,%d]\n", matches, wildchars, sequenceSize);

    // If not all nucleotide chars or less than 90% are G,A,T,C or N this is a protein query
    if (matches + wildchars < sequenceSize || matches < sequenceSize * 0.9)
    {
        return encoding_protein;
    }

    return encoding_nucleotide;
}

Here is the caller graph for this function:

void encoding_encodeSequence	(	char *	sequence,
		int4	sequenceSize,
		uint4	alphabetType
	)

Definition at line 505 of file encoding.c.

References wildcardEdit::code, encoding_codesArray, and int4.

Referenced by dust_dustSequence(), and main().

{
        int4 count;
    unsigned char code;

    // Convert letters to codes
    count = 0;
    while (count < sequenceSize)
    {
        code = encoding_codesArray[(unsigned char)(sequence[count])];
        sequence[count] = code;
        count++;
    }
}

Here is the caller graph for this function:

void encoding_free

(

)

Definition at line 521 of file encoding.c.

References wildcardEdit::code, encoding_alphabetType, encoding_codesArray, encoding_complements, encoding_nucleotide, encoding_numLetters, encoding_packedByteLookup, and encoding_wildcards.

Referenced by main().

{
        unsigned char code;

        free(encoding_codesArray);
        free(encoding_packedByteLookup);

    if (encoding_alphabetType == encoding_nucleotide)
        free(encoding_complements);

    code = 0;
    while (code < encoding_numLetters)
    {
        // Declare memory for regular codes
        free(encoding_wildcards[code].replacementCodes);
                code++;
    }
}

Here is the caller graph for this function:

unsigned char encoding_getCode

(

unsigned char

letter

)

Definition at line 215 of file encoding.c.

References encoding_codesArray.

Referenced by encoding_initialize(), PSSMatrix_create(), scoreMatrix_load(), and wildcards_initialize().

{
        return encoding_codesArray[letter];
}

Here is the caller graph for this function:

unsigned char encoding_getComplement

(

unsigned char

code

)

Definition at line 209 of file encoding.c.

References encoding_complements.

Referenced by print_constructAlignment(), and PSSMatrix_create().

{
        return encoding_complements[code];
}

Here is the caller graph for this function:

unsigned char encoding_getLetter

(

unsigned char

code

)

Definition at line 221 of file encoding.c.

References encoding_lettersArray, and encoding_numCodes.

Referenced by chooseWilds_printOccurenceMatrix(), encoding_printLetters(), nucleotideLookup_print(), print_constructAlignment(), print_singleSequence(), scoreMatrix_print(), wildcards_outputWildcards(), wildcards_printWildcard(), wordLookupDFA_print(), and wordLookupDFA_printCodes().

{
        if (code < encoding_numCodes)
                return encoding_lettersArray[code];
        else
                return '_';
}

Here is the caller graph for this function:

void encoding_initialize

(

unsigned char

alphabetType

)

Definition at line 97 of file encoding.c.

References packedByte::codes, encoding_aaLettersArray, encoding_aaNumCodes, encoding_aaNumLetters, encoding_aaNumRegularLetters, encoding_aaSentinalCode, encoding_aaUnknownCode, encoding_aaWildcards, encoding_alphabetType, encoding_codesArray, encoding_complements, encoding_getCode(), encoding_lettersArray, encoding_nucleotide, encoding_nuComplementsArray, encoding_nuLettersArray, encoding_numCodes, encoding_numLetters, encoding_numRegularLetters, encoding_nuNumCodes, encoding_nuNumLetters, encoding_nuNumRegularLetters, encoding_nuSentinalCode, encoding_nuUnknownCode, encoding_nuWildcards, encoding_packedByteLookup, encoding_protein, encoding_sentinalCode, encoding_unknownCode, encoding_wildcards, global_malloc(), int4, wildcard::letter, wildcard::replacementCodes, wildcard::replacementCounter, and wildcard::replacements.

Referenced by main().

{
        unsigned char letter = 0, code = 0, count;
    int4 packedByte;

    // Record alphabet type
    encoding_alphabetType = alphabetType;

    if (encoding_alphabetType == encoding_protein)
    {
        // Use amino acid alphabet
        encoding_lettersArray = encoding_aaLettersArray;
        encoding_unknownCode = encoding_aaUnknownCode;
        encoding_sentinalCode = encoding_aaSentinalCode;
        encoding_numLetters = encoding_aaNumLetters;
        encoding_numRegularLetters = encoding_aaNumRegularLetters;
        encoding_numCodes = encoding_aaNumCodes;
                encoding_wildcards = encoding_aaWildcards;
    }
    else
    {
        // Use nucleotide alphabet
        encoding_lettersArray = encoding_nuLettersArray;
        encoding_unknownCode = encoding_nuUnknownCode;
        encoding_sentinalCode = encoding_nuSentinalCode;
        encoding_numLetters = encoding_nuNumLetters;
        encoding_numRegularLetters = encoding_nuNumRegularLetters;
        encoding_numCodes = encoding_nuNumCodes;
                encoding_wildcards = encoding_nuWildcards;

        // Construct byte unpack table
                encoding_packedByteLookup = (struct packedByte*)global_malloc(sizeof(struct packedByte) * 256);

        // For each packed byte
        packedByte = 0;
        while (packedByte < 256)
        {
                // Store unpacked codes
                        encoding_packedByteLookup[packedByte].codes[0] = (packedByte >> 6) & 0x3;
                        encoding_packedByteLookup[packedByte].codes[1] = (packedByte >> 4) & 0x3;
                        encoding_packedByteLookup[packedByte].codes[2] = (packedByte >> 2) & 0x3;
                        encoding_packedByteLookup[packedByte].codes[3] = packedByte & 0x3;

            packedByte++;
        }
    }

        // Create array for translating amino acid characters into the related byte code
        encoding_codesArray = (unsigned char*)global_malloc(256);

        // Iterate for each possible character value
        while (letter < 255)
        {
                // By default use unknown code (eg. U or N)
                encoding_codesArray[letter] = encoding_unknownCode;

                // If a valid code (found by looking up lettersArray)
                code = 0;
                while (code < encoding_numCodes)
                {
                        if (toupper(letter) == encoding_lettersArray[code])
                        {
                                // Add to codes array
                                encoding_codesArray[letter] = code;
                        }
                        code++;
                }
                letter++;
        }

    // Initialize array for converting wildcards to regular letters
    code = 0;
    while (code < encoding_numLetters)
    {
        // Declare memory for regular codes
        encoding_wildcards[code].replacementCodes = (char*)global_malloc(sizeof(char) *
                                                    encoding_wildcards[code].numCodes);

                // Initialize replacement counter
        encoding_wildcards[code].replacementCounter = 0;

        // For each regular letter
        count = 0;
        while (count < encoding_wildcards[code].numCodes)
        {
            // Convert to encoded form
            encoding_wildcards[code].replacementCodes[count]
                = encoding_codesArray[encoding_wildcards[code].replacements[count]];

            count++;
        }
        code++;
    }

    // Construct complements array
    if (encoding_alphabetType == encoding_nucleotide)
    {
        encoding_complements = (unsigned char*)global_malloc(sizeof(unsigned char) * encoding_numLetters);

        // For each code
        code = 0;
        while (code < encoding_numLetters)
        {
                // Store the code of the complement
                        encoding_complements[code] = encoding_getCode(encoding_nuComplementsArray[code]);

                code++;
        }
    }
}

Here is the call graph for this function:

Here is the caller graph for this function:

void encoding_insertWilds	(	unsigned char *	subject,
		unsigned char *	edits,
		unsigned char *	endEdits
	)			[inline]

Definition at line 299 of file encoding.c.

References uint4, and vbyte_getVbyte.

Referenced by search_nucleotideSsearch().

{
    uint4 wildcardPosition;
        unsigned char wildcard;

    // For each edit
    while (edits < endEdits)
    {
        // Read wildcard
                wildcard = *edits;
        edits++;

        // Read its position
                vbyte_getVbyte(edits, &wildcardPosition);

        // Add wildcard
        subject[wildcardPosition] = wildcard;
    }
}

Here is the caller graph for this function:

void encoding_printLetters	(	unsigned char	code,
		int4	numLetters
	)

Definition at line 491 of file encoding.c.

References encoding_getLetter().

Referenced by ungappedExtension_nucleotideExtend().

{
        if (numLetters > 0)
                printf("%c", encoding_getLetter((code >> 6) & 0x3));
        if (numLetters > 1)
                printf("%c", encoding_getLetter((code >> 4) & 0x3));
        if (numLetters > 2)
                printf("%c", encoding_getLetter((code >> 2) & 0x3));
        if (numLetters > 3)
                printf("%c", encoding_getLetter(code & 0x3));
        fflush(stdout);
}

Here is the call graph for this function:

Here is the caller graph for this function:

unsigned char encoding_randomEncodedLetter

(

unsigned char

code

)

Definition at line 277 of file encoding.c.

References encoding_alphabetType, encoding_protein, encoding_wildcards, wildcard::replacementCodes, and wildcard::replacementCounter.

Referenced by dust_dustSequence(), encoding_replaceWildcards(), and PSSMatrix_create().

{
    // For protein alphabet use robinson&robinson frequencies
        if (encoding_alphabetType == encoding_protein)
    {
        encoding_wildcards[code].replacementCounter++;
        if (encoding_wildcards[code].replacementCounter >= encoding_wildcards[code].numCodes)
                        encoding_wildcards[code].replacementCounter = 0;

        return encoding_wildcards[code].replacementCodes[encoding_wildcards[code].replacementCounter];
    }
    else
    {
                encoding_wildcards[code].replacementCounter++;
        if (encoding_wildcards[code].replacementCounter >= encoding_wildcards[code].numCodes)
                        encoding_wildcards[code].replacementCounter = 0;

        return encoding_wildcards[code].replacementCodes[encoding_wildcards[code].replacementCounter];
        }
}

Here is the caller graph for this function:

int4 encoding_replaceWildcards	(	struct memSingleBlock *	wildcardEdits,
		unsigned char *	sequence,
		int4	sequenceSize
	)

Definition at line 439 of file encoding.c.

References wildcardEdit::code, encoding_numRegularLetters, encoding_randomEncodedLetter(), int4, memSingleBlock_newEntry(), memSingleBlock::numEntries, and wildcardEdit::position.

Referenced by main().

{
        struct wildcardEdit* wildcardEdit;
        int4 count = 0;

    // Initialize list of wildcard edits
    wildcardEdits->numEntries = 0;

    // First scan through sequence and replace wilds
        while (count < sequenceSize)
    {
        // If a wild
        if (sequence[count] >= encoding_numRegularLetters)
        {
                // Record location and code of wild
                        wildcardEdit = memSingleBlock_newEntry(wildcardEdits);
            wildcardEdit->code = sequence[count];
            wildcardEdit->position = count;

            // Code replacement
            sequence[count] = encoding_randomEncodedLetter(sequence[count]);
        }
        count++;
        }

    return wildcardEdits->numEntries;
}

Here is the call graph for this function:

Here is the caller graph for this function:

---

Variable Documentation

unsigned char encoding_alphabetType

Definition at line 19 of file encoding.c.

Referenced by alignments_createNew(), alignments_findGoodAlignments(), blast_search(), encoding_free(), encoding_initialize(), encoding_randomEncodedLetter(), gappedScoring_score(), hitMatrix_free(), hitMatrix_initialize(), hitMatrix_reinitialize(), main(), parameters_free(), print_constructAlignment(), print_gappedExtension(), print_XMLheader(), PSSMatrix_create(), PSSMatrix_free(), readdb_nextVolume(), statistics_calculateUngappedKarlinParameters(), statistics_initialize(), ungappedExtension_findSeed(), unpack_free(), unpack_loadSubject(), unpack_unpackSubject(), wordLookupDFA_build(), and wordLookupDFA_print().

char* encoding_alphabetTypes[2]

Definition at line 23 of file encoding.c.

Referenced by main().

unsigned char* encoding_codesArray

Definition at line 15 of file encoding.c.

Referenced by encoding_encodeSequence(), encoding_free(), encoding_getCode(), and encoding_initialize().

unsigned char* encoding_lettersArray

Definition at line 16 of file encoding.c.

Referenced by encoding_getLetter(), and encoding_initialize().

unsigned char encoding_numCodes

Definition at line 18 of file encoding.c.

Referenced by cluster_buildPSSM(), encoding_getLetter(), encoding_initialize(), PSSMatrix_print(), scoreMatrix_create(), scoreMatrix_load(), and scoreMatrix_print().

unsigned char encoding_numLetters

Definition at line 18 of file encoding.c.

Referenced by chooseWilds_printOccurenceMatrix(), cluster_averageWildcodeScore(), encoding_free(), encoding_initialize(), main(), scoreMatrix_load(), wildcards_averageResidueWildMatch(), wildcards_countBits(), wildcards_initializeCountOccurences(), wildcards_outputWildcards(), wildcards_printWildcard(), wildcards_scoreCandidates(), and wildcards_scoreResidueWildMatch().

unsigned char encoding_numRegularLetters

Definition at line 18 of file encoding.c.

Referenced by dust_dustSequence(), encoding_initialize(), encoding_replaceWildcards(), index_processQuery(), main(), PSSMatrix_create(), scoreMatrix_create(), statistics_calculateUngappedKarlinParameters(), and wordLookupDFA_getNeighbours().

struct packedByte* encoding_packedByteLookup

Definition at line 13 of file encoding.c.

Referenced by encoding_byteUnpackRegion(), encoding_free(), and encoding_initialize().

unsigned char encoding_sentinalCode

Definition at line 17 of file encoding.c.

Referenced by encoding_initialize(), main(), PSSMatrix_create(), readdb_getChildren(), scoreMatrix_create(), scoreMatrix_load(), writedb_addSequence(), writedb_close(), and writedb_initialize().

unsigned char encoding_unknownCode

Definition at line 17 of file encoding.c.

Referenced by encoding_initialize().

struct wildcard* encoding_wildcards

Definition at line 20 of file encoding.c.

Referenced by encoding_free(), encoding_initialize(), encoding_randomEncodedLetter(), and scoreMatrix_create().

---