int main(int argc, char *argv[]) {
#ifndef _OPENMP
fprintf(stderr, "\nERROR: Program built with compiler lacking OpenMP support.\n");
fprintf(stderr, "See SEAStAR README file for information about suitable compilers.\n");
exit(EXIT_FAILURE);
#endif
///////////////////////////
// Variable declarations
///////////////////////////
// Input filenames
UT_string *in_read1_fq_fn, *in_read2_fq_fn, *in_single1_fq_fn, *in_single2_fq_fn;
utstring_new(in_read1_fq_fn);
utstring_new(in_read2_fq_fn);
utstring_new(in_single1_fq_fn);
utstring_new(in_single2_fq_fn);
// Output filenames
UT_string *out_read1_fn, *out_read2_fn, *out_single1_fn, *out_single2_fn, *out_mates_fn, *out_filetype;
utstring_new(out_filetype);
utstring_new(out_read1_fn);
utstring_new(out_read2_fn);
utstring_new(out_single1_fn);
utstring_new(out_single2_fn);
utstring_new(out_mates_fn);
// Read name prefix
UT_string *out_read_prefix;
utstring_new(out_read_prefix);
// Flags
int singles_flag = 0; // 1 when two output singles files being written
int num_input_singles_files = 0;
// Read counters
unsigned long int mp_org = 0, R1_org = 0, R2_org = 0, singlet1_org = 0, singlet2_org = 0;
unsigned long int mp_cnt = 0, R1_cnt = 0, R2_cnt = 0, singlet1_cnt = 0, singlet2_cnt = 0, s1_cnt = 0, s2_cnt = 0;
unsigned long int comp_r1 = 0, comp_r2 = 0, comp_s1 = 0, comp_s2 = 0;
unsigned long int read1_singlet_cnt = 0, read2_singlet_cnt = 0;
////////////////////////////////////////////////////////////////////////
// All done with variable declarations!!
///////////////////////////////////
// Command line argtable settings
///////////////////////////////////
struct arg_lit *gzip = arg_lit0("z", "gzip", "Output converted files in gzip compressed format. [NULL]");
struct arg_lit *inv_singles = arg_lit0("v", "invert_singles", "Causes singles output to be the inverse of the input. 2->1 or 1->2 [NULL]");
struct arg_lit *num_singles = arg_lit0("s", "singles", "Write two singlet files, one for each mate-paired input file. [NULL]");
struct arg_rem *sing_rem = arg_rem(NULL, "Note! -v is only valid when there are input singlet reads. -s is only valid when there are NO input singlet reads.");
struct arg_str *pre_read_id = arg_str0(NULL, "prefix", "<string>", "Prefix to add to read identifiers. [out_prefix]");
struct arg_lit *no_pre = arg_lit0(NULL, "no_prefix", "Do not change the read names in any way. [NULL]");
struct arg_lit *pre_read_len = arg_lit0(NULL, "add_len", "Add the final trimmed length value to the read id prefix. [length not added]");
struct arg_dbl *prob = arg_dbl0("p","correct_prob","<d>","Probability that output reads are correct. 0.0 disables quality trimming. [0.5]");
struct arg_int *fixed_len = arg_int0("f","fixed_len","<u>","Trim all reads to a fixed length, still filtering on quality [no fixed length]");
struct arg_int *len = arg_int0("l","min_read_len","<u>","Minimum length of a singlet or longest-mate in nucleotides [24]");
struct arg_int *mate_len = arg_int0("m","min_mate_len","<u>","Minimum length of the shortest mate in nucleotides [min_read_len]");
struct arg_dbl *entropy = arg_dbl0("e","entropy_filter","<d>","Remove reads with per position information below given value (in bits per dinucleotide) [No filter]");
struct arg_lit *entropy_strict = arg_lit0(NULL, "entropy_strict", "Reject reads for low entropy overall, not just the retained part after trimming [NULL]");
struct arg_lit *mates = arg_lit0(NULL, "mates_file", "Produce a Velvet compatible interleaved paired read output file (e.g. <out_prefix>_mates.fastq). [NULL]");
struct arg_lit *no_rev = arg_lit0(NULL, "no_rev", "By default, the second read in each pair is reversed for colorspace --mate-file output. --no_rev disables reversing. [rev]");
struct arg_lit *only_mates = arg_lit0(NULL, "only_mates", "Supress writing .read1 and .read2 outputs. Requires --mates_file. [NULL]");
struct arg_lit *fasta = arg_lit0(NULL, "fasta", "Write FASTA format files instead of FASTQ for all outputs (e.g. <out_prefix>.<read_type>.fasta). [FASTQ]");
struct arg_file *input = arg_file1(NULL, NULL, "<in_prefix>", "Input file prefix: (e.g. <in_prefix>_single.fastq [<in_prefix>_read1.fastq <in_prefix>_read2.fastq]) ");
struct arg_file *output = arg_file1(NULL, NULL, "<out_prefix>", "Output file prefix: (e.g. <out_prefix>_single.fastq [<out_prefix>_read1.fastq <out_prefix>_read2.fastq]) ");
struct arg_lit *version = arg_lit0(NULL,"version","Print the build version and exit.");
struct arg_lit *h = arg_lit0("h", "help", "Request help.");
struct arg_end *end = arg_end(20);
void *argtable[] = {h,version,gzip,inv_singles,num_singles,sing_rem,prob,len,mate_len,fixed_len,pre_read_id,pre_read_len,no_pre,entropy,entropy_strict,mates,no_rev,only_mates,fasta,input,output,end};
int arg_errors = 0;
////////////////////////////////////////////////////////////////////////
// Handle command line processing (via argtable2 library)
////////////////////////////////////////////////////////////////////////
arg_errors = arg_parse(argc, argv, argtable);
if (version->count) {
fprintf(stderr, "%s version: %s\n", argv[0], SS_BUILD_VERSION);
exit(EXIT_SUCCESS);
}
if (h->count) {
fprintf(stderr,"\ntrim_fastq is a utility for performing quality and information-based\n");
fprintf(stderr,"trimming on paired or unpaired, nucleotide or SOLiD colorspace reads. \n\n");
arg_print_syntaxv(stderr, argtable, "\n\n");
arg_print_glossary(stderr, argtable, "%-25s %s\n");
fprintf(stderr, "\nInput and output \"prefixes\" are the part of the filename before:\n");
fprintf(stderr, "_single.fastq [_read1.fastq _read2.fastq] A singlets (single) file\n");
fprintf(stderr, "is required. Mate-paired read files are automatically used if present.\n");
fprintf(stderr, "Multiple output files only produced for mate-paired inputs.\n");
fprintf(stderr, "\nNote! Input and output files may be gzipped, and outputs can be written\n");
fprintf(stderr, "as either FASTQ or FASTA format files.\n");
exit(EXIT_FAILURE);
}
if (arg_errors) {
arg_print_errors(stderr, end, "trimfastq");
arg_print_syntaxv(stderr, argtable, "\n");
exit(EXIT_FAILURE);
}
// Validate entropy
if (entropy->count) {
entropy_cutoff = entropy->dval[0];
if ((entropy_cutoff < 0.0) || (entropy_cutoff > 4.0)) {
fprintf(stderr, "entropy_filter must be [0.0 - 4.0] \n");
exit(EXIT_FAILURE);
}
strict_ent = entropy_strict->count;
} else {
if (entropy_strict->count) {
fprintf(stderr, "Error: --entropy_strict requires --entropy_filter.\n");
exit(EXIT_FAILURE);
}
entropy_cutoff = -1.0;
}
// Validate error_prob
if (prob->count) {
err_prob = prob->dval[0];
if ((err_prob < 0.0) || (err_prob > 1.0)) {
fprintf(stderr, "--correct_prob (-p) must be 0.0 - 1.0 inclusive\n");
exit(EXIT_FAILURE);
}
} else {
err_prob = 0.5;
}
// Validate min read len
if (len->count) {
min_len = len->ival[0];
if (min_len <= 0) {
fprintf(stderr, "min_read_len must be > 0\n");
exit(EXIT_FAILURE);
}
} else {
min_len = 24;
}
// Validate min mate len
if (mate_len->count) {
min_mate_len = mate_len->ival[0];
if (min_mate_len <= 0) {
fprintf(stderr, "min_mate_len must be > 0\n");
exit(EXIT_FAILURE);
}
if (min_mate_len > min_len) {
fprintf(stderr, "min_mate_len must be <= min_len\n");
exit(EXIT_FAILURE);
}
} else {
min_mate_len = min_len;
}
if (fixed_len->count) {
fix_len = min_mate_len = min_len = fixed_len->ival[0];
if ((mate_len->count) || (len->count)) {
fprintf(stderr, "fixed_len cannot be used with min_read_len or min_mate_len\n");
exit(EXIT_FAILURE);
}
if (fix_len <= 0) {
fprintf(stderr, "fixed_len must be > 0\n");
exit(EXIT_FAILURE);
}
} else {
fix_len = 0;
}
if (pre_read_id->count) {
if (no_pre->count) {
fprintf(stderr, "Error: Both --prefix and --no_prefix were specified.\n");
exit(EXIT_FAILURE);
}
if (! strlen(pre_read_id->sval[0])) {
fprintf(stderr, "Read ID prefix may not be zero length.\n");
exit(EXIT_FAILURE);
}
if (strchr(pre_read_id->sval[0], ':') || strchr(pre_read_id->sval[0], '|') || strchr(pre_read_id->sval[0], '+') || strchr(pre_read_id->sval[0], '/')) {
fprintf(stderr, "Read ID prefix '%s' may not contain the characters ':', '|', '+' or '/'.\n", pre_read_id->sval[0]);
exit(EXIT_FAILURE);
}
// Build default read ID prefix
ss_strcat_utstring(out_read_prefix, pre_read_id->sval[0]);
} else {
if (!no_pre->count) {
if (strchr(output->filename[0], ':') || strchr(output->filename[0], '|') || strchr(output->filename[0], '+') || strchr(output->filename[0], '/')) {
fprintf(stderr, "Read ID prefix '%s' (from output prefix) may not contain the characters ':', '|', '+' or '/'.\n", output->filename[0]);
fprintf(stderr, "Hint: Use the --prefix parameter if the output file prefix contains path information.\n");
exit(EXIT_FAILURE);
}
// Build default read ID prefix
ss_strcat_utstring(out_read_prefix, output->filename[0]);
}
}
if ((only_mates->count) && (!mates->count)) {
fprintf(stderr, "--only_mates requires --mates.\n");
exit(EXIT_FAILURE);
}
if ((no_rev->count) && (!mates->count)) {
fprintf(stderr, "--no_rev requires --mates.\n");
exit(EXIT_FAILURE);
}
// Check for null string prefixes
if (!(strlen(input->filename[0]) && strlen(output->filename[0]))) {
fprintf(stderr, "Error: NULL prefix strings are not permitted.\n");
exit(EXIT_FAILURE);
}
// Construct input filenames
utstring_printf(in_read1_fq_fn, "%s.read1.fastq", input->filename[0]);
utstring_printf(in_read2_fq_fn, "%s.read2.fastq", input->filename[0]);
utstring_printf(in_single1_fq_fn, "%s.single.fastq", input->filename[0]);
FILE *in_read_file = NULL;
num_input_singles_files = 1;
// Try to open a singlet fastq file
// Check singlet output options -s and -v
// Set input singlet names to
// - *.single.fastq or
// - *.single1.fastq and *.single2.fastq
if (!(in_read_file = ss_get_gzFile(utstring_body(in_single1_fq_fn), "r"))) {
utstring_clear(in_single1_fq_fn);
utstring_printf(in_single1_fq_fn, "%s.single1.fastq", input->filename[0]);
utstring_printf(in_single2_fq_fn, "%s.single2.fastq", input->filename[0]);
num_input_singles_files = 2;
if ((in_read_file = ss_get_gzFile(utstring_body(in_single1_fq_fn), "r")) || (in_read_file = ss_get_gzFile(utstring_body(in_single2_fq_fn), "r"))) {
singles_flag = 1; // Two singlet outputs
} else {
singles_flag = num_singles->count; // Number of singlet outputs set by -s parm
if (inv_singles->count) {
fprintf(stderr, "Error: Invalid option -v, No input singlet file(s) found. Use -s to select multiple output singlet files.\n");
exit(EXIT_FAILURE);
}
}
}
if (in_read_file) {
gzclose(in_read_file);
if (num_singles->count) {
fprintf(stderr, "Error: Invalid option -s, Input singlet file(s) found, use -v to change the number of output singlet files.\n");
exit(EXIT_FAILURE);
}
}
// singles->count inverts the current singles file input scheme
singles_flag = (singles_flag ^ inv_singles->count);
// Check if input fastq is colorspace
// If some files are colorspace and some are basespace, throw an error
int fcount = 0;
int cscount = 0;
fcount += ss_is_fastq(utstring_body(in_read1_fq_fn));
fcount += ss_is_fastq(utstring_body(in_read2_fq_fn));
fcount += ss_is_fastq(utstring_body(in_single1_fq_fn));
fcount += ss_is_fastq(utstring_body(in_single2_fq_fn));
cscount += (ss_is_fastq(utstring_body(in_read1_fq_fn)) && ss_is_colorspace_fastq(utstring_body(in_read1_fq_fn)));
cscount += (ss_is_fastq(utstring_body(in_read2_fq_fn)) && ss_is_colorspace_fastq(utstring_body(in_read2_fq_fn)));
cscount += (ss_is_fastq(utstring_body(in_single1_fq_fn)) && ss_is_colorspace_fastq(utstring_body(in_single1_fq_fn)));
cscount += (ss_is_fastq(utstring_body(in_single2_fq_fn)) && ss_is_colorspace_fastq(utstring_body(in_single2_fq_fn)));
if (cscount && (cscount != fcount)) {
printf("Error: Mixed colorspace and basespace FASTQ files detected\n");
exit(EXIT_FAILURE);
}
colorspace_flag = cscount ? 1 : 0;
// Output filenames
if (fasta->count) {
ss_strcat_utstring(out_filetype, "fasta");
read_count_divisor = 2;
} else {
ss_strcat_utstring(out_filetype, "fastq");
read_count_divisor = 4;
}
if (!only_mates->count) {
utstring_printf(out_read1_fn, "%s.read1.%s", output->filename[0], utstring_body(out_filetype));
utstring_printf(out_read2_fn, "%s.read2.%s", output->filename[0], utstring_body(out_filetype));
}
if (singles_flag == 1) {
utstring_printf(out_single1_fn, "%s.single1.%s", output->filename[0], utstring_body(out_filetype));
utstring_printf(out_single2_fn, "%s.single2.%s", output->filename[0], utstring_body(out_filetype));
} else {
utstring_printf(out_single1_fn, "%s.single.%s", output->filename[0], utstring_body(out_filetype));
}
if (mates->count) {
utstring_printf(out_mates_fn, "%s.mates.%s", output->filename[0], utstring_body(out_filetype));
}
////////////////////////////////////////////////////////////////////////////////////////////////
// Begin processing!
#ifdef _OPENMP
omp_set_num_threads(10);
#endif
// This is the value of a non-valid pipe descriptor
#define NO_PIPE 0
int r1_pipe[2];
int r2_pipe[2];
int s1_pipe[2];
int s2_pipe[2];
pipe(r1_pipe);
pipe(r2_pipe);
pipe(s1_pipe);
pipe(s2_pipe);
int r1_out_pipe[2];
int r2_out_pipe[2];
int mates_out_pipe[2];
int s1_out_pipe[2];
int s2_out_pipe[2];
pipe(r1_out_pipe);
pipe(r2_out_pipe);
pipe(mates_out_pipe);
pipe(s1_out_pipe);
pipe(s2_out_pipe);
#pragma omp parallel sections default(shared)
{
#pragma omp section
{ // Read1 reader
fq_stream_trimmer(in_read1_fq_fn, r1_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_r1, &R1_org, '\0', fasta->count);
}
#pragma omp section
{ // Read1 writer
R1_cnt = ss_stream_writer(out_read1_fn, r1_out_pipe[0], gzip->count) / read_count_divisor;
}
#pragma omp section
{ // Read2 reader
fq_stream_trimmer(in_read2_fq_fn, r2_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_r2, &R2_org, '\0', fasta->count);
}
#pragma omp section
{ // Read2 writer
R2_cnt = ss_stream_writer(out_read2_fn, r2_out_pipe[0], gzip->count) / read_count_divisor;
}
#pragma omp section
{ // Single1 reader
// When there is only one input singles file, but two output singles files, then supply which mate to use for this stream in the split parameter
if ((singles_flag) && (num_input_singles_files == 1)) {
singlet1_cnt = fq_stream_trimmer(in_single1_fq_fn, s1_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_s1, &singlet1_org, '1', fasta->count);
} else {
singlet1_cnt = fq_stream_trimmer(in_single1_fq_fn, s1_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_s1, &singlet1_org, '\0', fasta->count);
}
}
#pragma omp section
{ // Single1 writer
s1_cnt = ss_stream_writer(out_single1_fn, s1_out_pipe[0], gzip->count) / read_count_divisor;
}
#pragma omp section
{ // Single2 reader
// When there is only one input singles file, but two output singles files, then supply which mate to use for this stream in the split parameter
if ((singles_flag) && (num_input_singles_files == 1)) {
singlet2_cnt = fq_stream_trimmer(in_single1_fq_fn, s2_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_s2, &singlet2_org, '2', fasta->count);
} else {
singlet2_cnt = fq_stream_trimmer(in_single2_fq_fn, s2_pipe[1], out_read_prefix, no_pre->count, pre_read_len->count, &comp_s2, &singlet2_org, '\0', fasta->count);
}
}
#pragma omp section
{ // Single2 writer
s2_cnt = ss_stream_writer(out_single2_fn, s2_out_pipe[0], gzip->count) / read_count_divisor;
}
#pragma omp section
{ // Velvet mates writer
// Divide count by 2 because both R1 and R2 reads go through this writer
mp_cnt = ss_stream_writer(out_mates_fn, mates_out_pipe[0], gzip->count) / 2 / read_count_divisor;
}
#pragma omp section
{ // Dispatcher
// Allocate data buffer strings
UT_string *r1_data;
utstring_new(r1_data);
UT_string *r2_data;
utstring_new(r2_data);
UT_string *s1_data;
utstring_new(s1_data);
UT_string *s2_data;
utstring_new(s2_data);
UT_string *rev_tmp;
utstring_new(rev_tmp);
UT_string *rev_data;
utstring_new(rev_data);
// Pipes
FILE *r1_in = fdopen(r1_pipe[0],"r");
FILE *r2_in = fdopen(r2_pipe[0],"r");
FILE *s1_in = fdopen(s1_pipe[0],"r");
FILE *s2_in = fdopen(s2_pipe[0],"r");
FILE *mates_out = fdopen(mates_out_pipe[1],"w");
FILE *r1_out = fdopen(r1_out_pipe[1],"w");
FILE *r2_out = fdopen(r2_out_pipe[1],"w");
FILE *s1_out = fdopen(s1_out_pipe[1],"w");
FILE *s2_out = fdopen(s2_out_pipe[1],"w");
if (!singles_flag) {
fclose(s2_out);
s2_out = s1_out;
}
// Flags for data left in single files
int single1_hungry = 1;
int single2_hungry = 1;
// Handle read1 and read2 files
while (ss_get_utstring(r1_in, r1_data)) {
if (!ss_get_utstring(r2_in, r2_data)) {
fprintf(stderr, "Error: Input read1 and read2 files are not synced\n");
exit(EXIT_FAILURE);
}
if (keep_read(r1_data)) {
if (keep_read(r2_data)) {
// Output both read1 and read2
if (mates->count) {
if (only_mates->count) {
// Interleaved velvet output only
output_read(r1_data, NULL, NULL, r1_in, NULL, mates_out, fasta->count);
if (no_rev->count || !colorspace_flag) {
output_read(r2_data, NULL, NULL, r2_in, NULL, mates_out, fasta->count);
} else {
output_read(r2_data, rev_data, rev_tmp, r2_in, NULL, mates_out, fasta->count);
}
} else {
// Interleaved velvet output and normal read file output
output_read(r1_data, NULL, NULL, r1_in, r1_out, mates_out, fasta->count);
if (no_rev->count || !colorspace_flag) {
output_read(r2_data, NULL, NULL, r2_in, r2_out, mates_out, fasta->count);
} else {
output_read(r2_data, rev_data, rev_tmp, r2_in, r2_out, mates_out, fasta->count);
}
}
} else {
// No interleaved velvet output
output_read(r1_data, NULL, NULL, r1_in, r1_out, NULL, fasta->count);
output_read(r2_data, NULL, NULL, r2_in, r2_out, NULL, fasta->count);
}
} else {
// Discard read2, output read1 as singlet
output_read(r1_data, NULL, NULL, r1_in, s1_out, NULL, fasta->count);
read1_singlet_cnt++;
}
} else {
if (keep_read(r2_data)) {
// Discard read1, output read2 as singlet
output_read(r2_data, NULL, NULL, r2_in, s2_out, NULL, fasta->count);
read2_singlet_cnt++;
}
}
// Process reads from singles here to take advantage of
// parallelism
if (single1_hungry || single2_hungry) {
if (single1_hungry) {
if (ss_get_utstring(s1_in, s1_data)) {
if (keep_read(s1_data)) {
output_read(s1_data, NULL, NULL, s1_in, s1_out, NULL, fasta->count);
}
} else {
single1_hungry = 0;
}
}
if (single2_hungry) {
if (ss_get_utstring(s2_in, s2_data)) {
if (keep_read(s2_data)) {
output_read(s2_data, NULL, NULL, s2_in, s2_out, NULL, fasta->count);
}
} else {
single2_hungry = 0;
}
}
}
}
while (single1_hungry || single2_hungry) {
if (single1_hungry) {
if (ss_get_utstring(s1_in, s1_data)) {
if (keep_read(s1_data)) {
output_read(s1_data, NULL, NULL, s1_in, s1_out, NULL, fasta->count);
}
} else {
single1_hungry = 0;
}
}
if (single2_hungry) {
if (ss_get_utstring(s2_in, s2_data)) {
if (keep_read(s2_data)) {
output_read(s2_data, NULL, NULL, s2_in, s2_out, NULL, fasta->count);
}
} else {
single2_hungry = 0;
}
}
}
fclose(r1_in);
fclose(r2_in);
fclose(s1_in);
fclose(s2_in);
fclose(mates_out);
fclose(r1_out);
fclose(r2_out);
fclose(s1_out);
if (singles_flag) {
fclose(s2_out);
}
// Free buffers
utstring_free(r1_data);
utstring_free(r2_data);
utstring_free(s1_data);
utstring_free(s2_data);
utstring_free(rev_tmp);
utstring_free(rev_data);
}
}
if (!(R1_org+singlet1_org+singlet2_org)) {
fprintf(stderr, "ERROR! No reads found in input files, or input(s) not found.\n");
exit(EXIT_FAILURE);
}
if (R1_org != R2_org) {
fprintf(stderr, "\nWarning! read1 and read2 fastq files did not contain an equal number of reads. %lu %lu\n", R1_org, R2_org);
}
if ((R1_org + R2_org) && !(singlet1_cnt + singlet2_cnt)) {
fprintf(stderr, "\nWarning! read1/read2 files were processed, but no corresponding input singlets were found.\n");
}
if (entropy->count) {
printf("\nLow complexity reads discarded: Read1: %lu, Read2: %lu, Singlets: %lu %lu\n", comp_r1, comp_r2, comp_s1, comp_s2);
}
mp_org = R1_org;
if (!only_mates->count) {
mp_cnt = R1_cnt;
}
printf("\nMatepairs: Before: %lu, After: %lu\n", mp_org, mp_cnt);
printf("Singlets: Before: %lu %lu After: %lu %lu\n", singlet1_org, singlet2_org, s1_cnt, s2_cnt);
printf("Read1 singlets: %lu, Read2 singlets: %lu, Original singlets: %lu %lu\n", read1_singlet_cnt, read2_singlet_cnt, singlet1_cnt, singlet2_cnt);
printf("Total Reads Processed: %lu, Reads retained: %lu\n", 2*mp_org+singlet1_org+singlet2_org, 2*mp_cnt+s1_cnt+s2_cnt);
utstring_free(in_read1_fq_fn);
utstring_free(in_read2_fq_fn);
utstring_free(in_single1_fq_fn);
utstring_free(in_single2_fq_fn);
utstring_free(out_read1_fn);
utstring_free(out_read2_fn);
utstring_free(out_single1_fn);
utstring_free(out_single2_fn);
utstring_free(out_mates_fn);
utstring_free(out_filetype);
utstring_free(out_read_prefix);
exit(EXIT_SUCCESS);
}
////////////////////////////////////////////////////////////////////////////////
// correct_reads
//
// Correct the reads in the file 'fqf' using the data structure of trusted
// kmers 'trusted', matrix of nt->nt error rates 'ntnt_prob' and prior nt
// probabilities 'prior_prob'. 'starts' and 'counts' help openMP parallelize
// the read processing. If 'pairedend_code' is 0, the reads are not paired;
// if it's 1, this file is the first of a pair so print all reads and withold
// combining; if it's 2, the file is the second of a pair so print all reads
// and then combine both 1 and 2.
////////////////////////////////////////////////////////////////////////////////
static void correct_reads(string fqf, int pe_code, bithash * trusted, vector<streampos> & starts, vector<unsigned long long> & counts, double ntnt_prob[Read::max_qual][4][4], double prior_prob[4]) {
// output directory
struct stat st_file_info;
string path_suffix = split(fqf,'/').back();
string out_dir("."+path_suffix);
if(stat(out_dir.c_str(), &st_file_info) == 0) {
cerr << "Hidden temporary directory " << out_dir << " already exists and will be used" << endl;
} else {
if(mkdir(out_dir.c_str(), S_IRWXU) == -1) {
cerr << "Failed to create hidden temporary directory " << out_dir << endl;
exit(EXIT_FAILURE);
}
}
// collect stats
stats * thread_stats = new stats[omp_get_max_threads()];
unsigned int chunk = 0;
#pragma omp parallel //shared(trusted)
{
int tid = omp_get_thread_num();
// input
ifstream reads_in(fqf.c_str());
unsigned int tchunk;
string header,ntseq,mid,strqual,corseq;
int trim_length;
char* nti;
Read *r;
#pragma omp critical
tchunk = chunk++;
while(tchunk < starts.size()) {
reads_in.seekg(starts[tchunk]);
// output
string toutf(out_dir+"/");
stringstream tconvert;
tconvert << tchunk;
toutf += tconvert.str();
if(overwrite_temp || stat(toutf.c_str(), &st_file_info) == -1) {
ofstream reads_out(toutf.c_str());
//cout << toutf << endl;
// output log
string tlogf = toutf + ".log";
ofstream corlog_out;
if(out_log) {
corlog_out.open(tlogf.c_str());
}
unsigned long long tcount = 0;
while(getline(reads_in, header)) {
//cout << tid << " " << header << endl;
// get sequence
getline(reads_in, ntseq);
//cout << ntseq << endl;
// convert ntseq to iseq
vector<unsigned int> iseq;
for(int i = 0; i < ntseq.size(); i++) {
nti = strchr(nts, ntseq[i]);
iseq.push_back(nti - nts);
}
// get quality values
getline(reads_in,mid);
//cout << mid << endl;
getline(reads_in,strqual);
//cout << strqual << endl;
vector<int> untrusted;
if(iseq.size() < trim_t)
trim_length = 0;
else {
for(int i = 0; i < iseq.size()-k+1; i++) {
if(!trusted->check(&iseq[i])) {
untrusted.push_back(i);
}
}
trim_length = quick_trim(strqual, untrusted);
//trim_length = iseq.size();
}
// fix error reads
if(untrusted.size() > 0) {
r = new Read(header, &iseq[0], strqual, untrusted, trim_length);
corseq = r->correct(trusted, ntnt_prob, prior_prob);
// output read w/ trim and corrections
output_read(reads_out, corlog_out, pe_code, header, ntseq, mid, strqual, corseq, thread_stats[tid]);
delete r;
} else {
output_read(reads_out, corlog_out, pe_code, header, ntseq, mid, strqual, ntseq.substr(0,trim_length), thread_stats[tid]);
// output read as trimmed
/*
if(contrail_out)
reads_out << header << "\t" << ntseq.substr(0,trim_length) << endl;
else
reads_out << header << endl << ntseq.substr(0,trim_length) << endl << mid << endl << strqual.substr(0,trim_length) << endl;
*/
}
if(++tcount == counts[tchunk])
break;
}
reads_out.close();
}
#pragma omp critical
tchunk = chunk++;
}
reads_in.close();
}
// combine stats
for(int i = 1; i < omp_get_max_threads(); i++) {
thread_stats[0].validated += thread_stats[i].validated;
thread_stats[0].corrected += thread_stats[i].corrected;
thread_stats[0].trimmed += thread_stats[i].trimmed;
thread_stats[0].trimmed_only += thread_stats[i].trimmed_only;
thread_stats[0].removed += thread_stats[i].removed;
}
// print stats
int suffix_index = fqf.rfind(".");
string outf;
if(suffix_index == -1) {
outf = fqf+".stats.txt";
} else {
outf = fqf.substr(0,suffix_index+1) + "stats.txt";
}
ofstream stats_out(outf.c_str());
stats_out << "Validated: " << thread_stats[0].validated << endl;
stats_out << "Corrected: " << thread_stats[0].corrected << endl;
stats_out << "Trimmed: " << thread_stats[0].trimmed << endl;
stats_out << "Trimmed only: " << thread_stats[0].trimmed_only << endl;
stats_out << "Removed: " << thread_stats[0].removed << endl;
stats_out.close();
}
