Пример #1
0
//err_dep = 0.83
void bam_likes_init(){
  mod = errmod_init(1-ERR_DEP);
}
Пример #2
0
int main_ld(int argc, char *argv[])
{
    int chr;                  //! chromosome identifier
    int beg;                  //! beginning coordinate for analysis
    int end;                  //! end coordinate for analysis
    int ref;                  //! ref
    long num_windows;         //! number of windows
    std::string msg;          //! string for error message
    bam_plbuf_t *buf;         //! pileup buffer
    ldData t;

    // parse the command line options
    std::string region = t.parseCommandLine(argc, argv);

    // check input BAM file for errors
    t.checkBAM();

    // initialize the sample data structure
    t.bam_smpl_init();

    // add samples
    t.bam_smpl_add();

    // initialize error model
    t.em = errmod_init(1.0-0.83);

    // parse genomic region
    int k = bam_parse_region(t.h, region, &chr, &beg, &end);
    if (k < 0)
    {
        msg = "Bad genome coordinates: " + region;
        fatal_error(msg, __FILE__, __LINE__, 0);
    }

    // fetch reference sequence
    t.ref_base = faidx_fetch_seq(t.fai_file, t.h->target_name[chr], 0, 0x7fffffff, &(t.len));

    // calculate the number of windows
    if (t.flag & BAM_WINDOW)
        num_windows = ((end-beg)-1)/t.win_size;
    else
    {
        t.win_size = (end-beg);
        num_windows = 1;
    }

    // iterate through all windows along specified genomic region
    for (long cw=0; cw < num_windows; cw++)
    {

        // construct genome coordinate string
        std::string scaffold_name(t.h->target_name[chr]);
        std::ostringstream winc(scaffold_name);
        winc.seekp(0, std::ios::end);
        winc << ":" << beg+(cw*t.win_size)+1 << "-" << ((cw+1)*t.win_size)+(beg-1);
        std::string winCoord = winc.str();

        // initialize number of sites to zero
        t.num_sites = 0;

        // parse the BAM file and check if region is retrieved from the reference
        if (t.flag & BAM_WINDOW)
        {
            k = bam_parse_region(t.h, winCoord, &ref, &(t.beg), &(t.end));
            if (k < 0)
            {
                msg = "Bad window coordinates " + winCoord;
                fatal_error(msg, __FILE__, __LINE__, 0);
            }
        }
        else
        {
            ref = chr;
            t.beg = beg;
            t.end = end;
            if (ref < 0)
            {
                msg = "Bad scaffold name: " + region;
                fatal_error(msg, __FILE__, __LINE__, 0);
            }
        }

        // initialize nucdiv variables
        t.init_ld();

        // create population assignments
        t.assign_pops();

        // initialize pileup
        buf = bam_plbuf_init(make_ld, &t);

        // fetch region from bam file
        if ((bam_fetch(t.bam_in->x.bam, t.idx, ref, t.beg, t.end, buf, fetch_func)) < 0)
        {
            msg = "Failed to retrieve region " + region + " due to corrupted BAM index file";
            fatal_error(msg, __FILE__, __LINE__, 0);
        }

        // finalize pileup
        bam_plbuf_push(0, buf);

        // calculate linkage disequilibrium statistics
        ld_func fp[3] = {&ldData::calc_zns, &ldData::calc_omegamax, &ldData::calc_wall};
        (t.*fp[t.output])();

        // print results to stdout
        t.print_ld(chr);

        // take out the garbage
        t.destroy_ld();
        bam_plbuf_destroy(buf);
    }
    // end of window interation

    errmod_destroy(t.em);
    samclose(t.bam_in);
    bam_index_destroy(t.idx);
    t.bam_smpl_destroy();
    free(t.ref_base);

    return 0;
}