C++ Genome::startContig方法代码示例

    const Genome *
ReadFASTAGenome(
    const char *fileName,
    const char *pieceNameTerminatorCharacters,
    bool spaceIsAPieceNameTerminator,
    unsigned chromosomePaddingSize)
{
    //
    // We need to know a bound on the size of the genome before we create the Genome object.
    // A bound is the number of bytes in the FASTA file, because we store at most one base per
    // byte.  Get the file size to use for this bound.
    //
    _int64 fileSize = QueryFileSize(fileName);
    bool isValidGenomeCharacter[256];

    for (int i = 0; i < 256; i++) {
        isValidGenomeCharacter[i] = false;
    }

    isValidGenomeCharacter['A'] = isValidGenomeCharacter['T'] = isValidGenomeCharacter['C'] = isValidGenomeCharacter['G'] = isValidGenomeCharacter['N'] = true;
    isValidGenomeCharacter['a'] = isValidGenomeCharacter['t'] = isValidGenomeCharacter['c'] = isValidGenomeCharacter['g'] = isValidGenomeCharacter['n'] = true;

    FILE *fastaFile = fopen(fileName, "r");
    if (fastaFile == NULL) {
        WriteErrorMessage("Unable to open FASTA file '%s' (even though we already got its size)\n",fileName);
        return NULL;
    }

    int lineBufferSize = 0;
    char *lineBuffer;
 
    //
    // Count the chromosomes
    //
    unsigned nChromosomes = 0;

    while (NULL != reallocatingFgets(&lineBuffer,&lineBufferSize,fastaFile)) {
        if (lineBuffer[0] == '>') {
            nChromosomes++;
        }
    }
    rewind(fastaFile);

    Genome *genome = new Genome(fileSize + (nChromosomes+1) * (size_t)chromosomePaddingSize, fileSize + (nChromosomes+1) * (size_t)chromosomePaddingSize, chromosomePaddingSize, nChromosomes + 1);

    char *paddingBuffer = new char[chromosomePaddingSize+1];
    for (unsigned i = 0; i < chromosomePaddingSize; i++) {
        paddingBuffer[i] = 'n';
    }
    paddingBuffer[chromosomePaddingSize] = '\0';

    bool warningIssued = false;
    bool inAContig = false;

    while (NULL != reallocatingFgets(&lineBuffer, &lineBufferSize, fastaFile)) {
        if (lineBuffer[0] == '>') {
            inAContig = true;
            //
            // A new contig.  Add in the padding first.
            //
            genome->addData(paddingBuffer);

            //
            // Now supply the chromosome name.
            //
            if (NULL != pieceNameTerminatorCharacters) {
                for (int i = 0; i < strlen(pieceNameTerminatorCharacters); i++) {
                    char *terminator = strchr(lineBuffer+1, pieceNameTerminatorCharacters[i]);
                    if (NULL != terminator) {
                        *terminator = '\0';
                    }
                }
            }
            if (spaceIsAPieceNameTerminator) {
                char *terminator = strchr(lineBuffer, ' ');
                if (NULL != terminator) {
                    *terminator = '\0';
                }
                terminator = strchr(lineBuffer, '\t');
                if (NULL != terminator) {
                    *terminator = '\0';
                }
            }
            char *terminator = strchr(lineBuffer, '\n');
            if (NULL != terminator) {
                *terminator = '\0';
            }
            terminator = strchr(lineBuffer, '\r');
            if (NULL != terminator) {
                *terminator = '\0';
            }
            genome->startContig(lineBuffer+1);
        } else {
            if (!inAContig) {
                WriteErrorMessage("\nFASTA file doesn't beging with a contig name (i.e., the first line doesn't start with '>').\n");
                soft_exit(1);
            }

            //
            // Convert it to upper case and truncate the newline before adding it to the genome.
//.........这里部分代码省略.........

//
// Makes a copy of a Genome, but with only one of the sex chromosomes.
//
// The fate of the mitochondrion is that of the X chromosome.
//
    Genome *
Genome::copy(bool copyX, bool copyY, bool copyM) const
{
    Genome *newCopy = new Genome(getCountOfBases(),getCountOfBases(), chromosomePadding);

    if (NULL == newCopy) {
        WriteErrorMessage("Genome::copy: failed to allocate space for copy.\n");
        return NULL;
    }

    const Genome::Contig *currentContig = NULL;
    const Genome::Contig *nextContig = getContigAtLocation(0);

    unsigned offsetInReference = 0;
    while (offsetInReference < getCountOfBases()) {
        if (NULL != nextContig && offsetInReference >= nextContig->beginningOffset) {
            //
            // Start of a new contig.  See if we want to skip it.
            //
            currentContig = nextContig;
            nextContig = getNextContigAfterLocation(offsetInReference + 1);
            if ((!copyX && !strcmp(currentContig->name,"chrX")) ||
                (!copyY && !strcmp(currentContig->name,"chrY")) ||
                (!copyM && !strcmp(currentContig->name,"chrM"))) {
                //
                // Yes, skip over this contig.
                //
                nextContig = getNextContigAfterLocation(offsetInReference + 1);
                if (NULL == nextContig) {
                    //
                    // The chromosome that we're skipping was the last one, so we're done.
                    //
                    break;
                } else {
                    offsetInReference = nextContig->beginningOffset;
                    continue;
                }
            } // If skipping this chromosome

            newCopy->startContig(currentContig->name);
        } // If new contig beginning

        const size_t maxCopySize = 10000;
        char dataBuffer[maxCopySize + 1];

        unsigned amountToCopy = maxCopySize;
        if (nextContig && nextContig->beginningOffset < offsetInReference + amountToCopy) {
            amountToCopy = nextContig->beginningOffset - offsetInReference;
        }

        if (getCountOfBases() < offsetInReference + amountToCopy) {
            amountToCopy = getCountOfBases() - offsetInReference;
        }

        memcpy(dataBuffer,getSubstring(offsetInReference,amountToCopy), amountToCopy);
        dataBuffer[amountToCopy] = '\0';

        newCopy->addData(dataBuffer);

        offsetInReference += amountToCopy;
    }

    newCopy->fillInContigLengths();
    newCopy->sortContigsByName();
    return newCopy;
}