void align_scoring_load_pairwise(gzFile file, const char* file_path,
scoring_t* scoring, char case_sensitive)
{
StrBuf* sbuf = strbuf_new(200);
size_t read_length;
int line_num = 0;
char a, b;
int score;
int num_pairs_added = 0;
while((read_length = strbuf_reset_gzreadline(sbuf, file)) > 0)
{
strbuf_chomp(sbuf);
if(sbuf->end > 0 && sbuf->b[0] != '#' && // line is not empty, not comment
!string_is_all_whitespace(sbuf->b)) // and not whitespace
{
if(read_length < 5)
{
_loading_error("Too few column headings", file_path, line_num, 0);
}
if(isspace(sbuf->b[1]))
{
// split by whitespace
a = sbuf->b[0];
size_t char2_pos;
for(char2_pos = 1;
sbuf->b[char2_pos] != '\0' && isspace(sbuf->b[char2_pos]);
char2_pos++);
if(char2_pos+2 >= sbuf->end || !isspace(sbuf->b[char2_pos+1]))
{
_loading_error("Line too short", file_path, line_num, 0);
}
b = sbuf->b[char2_pos];
if(!parse_entire_int(sbuf->b+char2_pos+2, &score))
{
_loading_error("Invalid number", file_path, line_num, 0);
}
}
else
{
if(sbuf->b[1] != sbuf->b[3])
{
_loading_error("Inconsistent separators used", file_path, line_num, 0);
}
a = sbuf->b[0];
b = sbuf->b[2];
if(!parse_entire_int(sbuf->b + 4, &score))
{
_loading_error("Invalid number", file_path, line_num, 0);
}
}
if(!case_sensitive)
{
a = tolower(a);
b = tolower(b);
}
scoring_add_mutation(scoring, a, b, score);
num_pairs_added++;
}
line_num++;
}
strbuf_free(sbuf);
if(num_pairs_added == 0)
{
_loading_error("No pairs added from file (file empty?)",
file_path, line_num, 0);
}
}
cmdline_t* cmdline_new(int argc, char **argv, scoring_t *scoring,
enum SeqAlignCmdType cmd_type)
{
cmdline_t* cmd = calloc(1, sizeof(cmdline_t));
cmd->file_list_length = 0;
cmd->file_list_capacity = 256;
cmd->file_paths1 = malloc(sizeof(char*) * cmd->file_list_capacity);
cmd->file_paths2 = malloc(sizeof(char*) * cmd->file_list_capacity);
cmd->seq1 = cmd->seq2 = NULL;
// All values initially 0
// Store defaults
score_t defaults[4] = {scoring->match, scoring->mismatch,
scoring->gap_open, scoring->gap_extend};
if(argc == 1) usage(NULL);
// First run through arguments to set up case_sensitive and scoring system
// case sensitive needs to be dealt with first
// (is is used to construct hash table for swap_scores)
char scoring_set = 0, substitutions_set = 0, match_set = 0, mismatch_set = 0;
int argi;
for(argi = 1; argi < argc; argi++)
{
if(strcasecmp(argv[argi], "--help") == 0 ||
strcasecmp(argv[argi], "-help") == 0 ||
strcasecmp(argv[argi], "-h") == 0)
{
usage(NULL);
}
else if(strcasecmp(argv[argi], "--case_sensitive") == 0)
{
cmd->case_sensitive = 1;
}
else if(strcasecmp(argv[argi], "--scoring") == 0)
{
if(scoring_set)
{
usage("More than one scoring system specified - not permitted");
}
if(strcasecmp(argv[argi+1], "PAM30") == 0)
{
scoring_system_PAM30(scoring);
}
else if(strcasecmp(argv[argi+1], "PAM70") == 0)
{
scoring_system_PAM70(scoring);
}
else if(strcasecmp(argv[argi+1], "BLOSUM80") == 0)
{
scoring_system_BLOSUM80(scoring);
}
else if(strcasecmp(argv[argi+1], "BLOSUM62") == 0)
{
scoring_system_BLOSUM62(scoring);
}
else if(strcasecmp(argv[argi+1], "DNA_HYBRIDIZATION") == 0)
{
scoring_system_DNA_hybridization(scoring);
}
else {
usage("Unknown --scoring choice, not one of "
"PAM30|PAM70|BLOSUM80|BLOSUM62");
}
scoring_set = 1;
argi++; // took an argument
}
}
for(argi = 1; argi < argc; argi++)
{
if(argv[argi][0] == '-')
{
// strcasecmp does case insensitive comparison
if(strcasecmp(argv[argi], "--freestartgap") == 0)
{
if(cmd_type != SEQ_ALIGN_NW_CMD)
usage("--freestartgap only valid with Needleman-Wunsch");
scoring->no_start_gap_penalty = true;
}
else if(strcasecmp(argv[argi], "--freeendgap") == 0)
{
if(cmd_type != SEQ_ALIGN_NW_CMD)
usage("--freeendgap only valid with Needleman-Wunsch");
scoring->no_end_gap_penalty = true;
}
else if(strcasecmp(argv[argi], "--nogaps") == 0)
{
scoring->no_gaps_in_a = true;
scoring->no_gaps_in_b = true;
}
else if(strcasecmp(argv[argi], "--nogapsin1") == 0)
{
scoring->no_gaps_in_a = true;
}
else if(strcasecmp(argv[argi], "--nogapsin2") == 0)
{
scoring->no_gaps_in_b = true;
}
else if(strcasecmp(argv[argi], "--nomismatches") == 0)
{
scoring->no_mismatches = true;
}
else if(strcasecmp(argv[argi], "--case_sensitive") == 0)
{
// Already dealt with
//case_sensitive = true;
}
else if(strcasecmp(argv[argi], "--printseq") == 0)
{
if(cmd_type != SEQ_ALIGN_SW_CMD)
usage("--printseq only valid with Smith-Waterman");
cmd->print_seq = true;
}
else if(strcasecmp(argv[argi], "--printmatrices") == 0)
{
cmd->print_matrices = true;
}
else if(strcasecmp(argv[argi], "--printscores") == 0)
{
if(cmd_type != SEQ_ALIGN_NW_CMD)
usage("--printscores only valid with Needleman-Wunsch");
cmd->print_scores = true;
}
else if(strcasecmp(argv[argi], "--printfasta") == 0)
{
cmd->print_fasta = true;
}
else if(strcasecmp(argv[argi], "--pretty") == 0)
{
cmd->print_pretty = true;
}
else if(strcasecmp(argv[argi], "--colour") == 0)
{
cmd->print_colour = true;
}
else if(strcasecmp(argv[argi], "--zam") == 0)
{
if(cmd_type != SEQ_ALIGN_NW_CMD)
usage("--zam only valid with Needleman-Wunsch");
cmd->zam_stle_output = true;
}
else if(strcasecmp(argv[argi], "--stdin") == 0)
{
// Similar to --file argument below
cmdline_add_files(cmd, "", NULL);
cmd->interactive = true;
}
else if(argi == argc-1)
{
// All the remaining options take an extra argument
usage("Unknown argument without parameter: %s", argv[argi]);
}
else if(strcasecmp(argv[argi], "--scoring") == 0)
{
// This handled above
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--substitution_matrix") == 0)
{
gzFile sub_matrix_file = gzopen(argv[argi+1], "r");
if(sub_matrix_file == NULL) usage("Couldn't read: %s", argv[argi+1]);
align_scoring_load_matrix(sub_matrix_file, argv[argi+1],
scoring, cmd->case_sensitive);
gzclose(sub_matrix_file);
substitutions_set = true;
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--substitution_pairs") == 0)
{
gzFile sub_pairs_file = gzopen(argv[argi+1], "r");
if(sub_pairs_file == NULL) usage("Couldn't read: %s", argv[argi+1]);
align_scoring_load_pairwise(sub_pairs_file, argv[argi+1],
scoring, cmd->case_sensitive);
gzclose(sub_pairs_file);
substitutions_set = true;
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--minscore") == 0)
{
if(cmd_type != SEQ_ALIGN_SW_CMD)
usage("--minscore only valid with Smith-Waterman");
if(!parse_entire_int(argv[argi+1], &cmd->min_score))
usage("Invalid --minscore <score> argument (must be a +ve int)");
cmd->min_score_set = true;
argi++;
}
else if(strcasecmp(argv[argi], "--maxhits") == 0)
{
if(cmd_type != SEQ_ALIGN_SW_CMD)
usage("--maxhits only valid with Smith-Waterman");
if(!parse_entire_uint(argv[argi+1], &cmd->max_hits_per_alignment))
usage("Invalid --maxhits <hits> argument (must be a +ve int)");
cmd->max_hits_per_alignment_set = true;
argi++;
}
else if(strcasecmp(argv[argi], "--context") == 0)
{
if(cmd_type != SEQ_ALIGN_SW_CMD)
usage("--context only valid with Smith-Waterman");
if(!parse_entire_uint(argv[argi+1], &cmd->print_context))
usage("Invalid --context <c> argument (must be >= 0)");
argi++;
}
else if(strcasecmp(argv[argi], "--match") == 0)
{
if(!parse_entire_int(argv[argi+1], &scoring->match))
{
usage("Invalid --match argument ('%s') must be an int", argv[argi+1]);
}
match_set = true;
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--mismatch") == 0)
{
if(!parse_entire_int(argv[argi+1], &scoring->mismatch))
{
usage("Invalid --mismatch argument ('%s') must be an int", argv[argi+1]);
}
mismatch_set = true;
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--gapopen") == 0)
{
if(!parse_entire_int(argv[argi+1], &scoring->gap_open))
{
usage("Invalid --gapopen argument ('%s') must be an int", argv[argi+1]);
}
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--gapextend") == 0)
{
if(!parse_entire_int(argv[argi+1], &scoring->gap_extend))
{
usage("Invalid --gapextend argument ('%s') must be an int",
argv[argi+1]);
}
argi++; // took an argument
}
else if(strcasecmp(argv[argi], "--file") == 0)
{
cmdline_add_files(cmd, argv[argi+1], NULL);
argi++; // took an argument
}
// Remaining options take two arguments but check themselves
else if(strcasecmp(argv[argi], "--files") == 0)
{
if(argi >= argc-2)
{
usage("--files option takes 2 arguments");
}
else if(strcmp(argv[argi+1], "-") == 0 && strcmp(argv[argi+2], "-") == 0)
{
// Read both from stdin
cmdline_add_files(cmd, argv[argi+1], NULL);
}
else
{
cmdline_add_files(cmd, argv[argi+1], argv[argi+2]);
}
argi += 2; // took two arguments
}
else if(strcasecmp(argv[argi], "--wildcard") == 0)
{
int wildscore = 0;
if(argi == argc-2 || strlen(argv[argi+1]) != 1 ||
!parse_entire_int(argv[argi+2], &wildscore))
{
usage("--wildcard <w> <s> takes a single character and a number");
}
scoring_add_wildcard(scoring, argv[argi+1][0], wildscore);
argi += 2; // took two arguments
}
else usage("Unknown argument '%s'", argv[argi]);
}
else
{
if(argc - argi != 2) usage("Unknown options: '%s'", argv[argi]);
break;
}
}
if((match_set && !mismatch_set && !scoring->no_mismatches) ||
(!match_set && mismatch_set))
{
usage("--match --mismatch must both be set or neither set");
}
else if(substitutions_set && !match_set)
{
// if substitution table set and not match/mismatch
scoring->use_match_mismatch = 0;
}
if(scoring->use_match_mismatch && scoring->match < scoring->mismatch) {
usage("Match value should not be less than mismatch penalty");
}
// Cannot guarantee that we can perform a global alignment if nomismatches
// and nogaps is true
if(cmd_type == SEQ_ALIGN_NW_CMD && scoring->no_mismatches &&
(scoring->no_gaps_in_a || scoring->no_gaps_in_b))
{
usage("--nogaps.. --nomismatches cannot be used at together");
}
// Check for extra unused arguments
// and set seq1 and seq2 if they have been passed
if(argi < argc)
{
cmd->seq1 = argv[argi];
cmd->seq2 = argv[argi+1];
}
if(cmd->seq1 == NULL && cmd->file_list_length == 0)
{
usage("No input specified");
}
if(cmd->zam_stle_output &&
(cmd->print_pretty || cmd->print_scores ||
cmd->print_colour || cmd->print_fasta))
{
usage("Cannot use --printscore, --printfasta, --pretty or --colour with "
"--zam");
}
return cmd;
}
void read_thread_args_parse(struct ReadThreadCmdArgs *args,
int argc, char **argv,
const struct option *longopts, bool correct_cmd)
{
size_t i;
int tmp_thresh; // 0 => no calling, -1 => auto
CorrectAlnInput task = CORRECT_ALN_INPUT_INIT;
uint8_t fq_offset = 0;
size_t dump_seq_n = 0, dump_mp_n = 0; // how many times are -g -G specified
PathFileReader tmp_pfile;
CorrectAlnInputBuffer *inputs = &args->inputs;
// Arg parsing
char cmd[100];
char shortopts[300];
cmd_long_opts_to_short(longopts, shortopts, sizeof(shortopts));
int used = 1, c;
char *tmp_path;
// silence error messages from getopt_long
// opterr = 0;
while((c = getopt_long_only(argc, argv, shortopts, longopts, NULL)) != -1) {
cmd_get_longopt_str(longopts, c, cmd, sizeof(cmd));
switch(c) {
case 0: /* flag set */ break;
case 'h': cmd_print_usage(NULL); break;
case 'o':
if(args->out_ctp_path != NULL) cmd_print_usage(NULL);
args->out_ctp_path = optarg;
break;
case 'p':
tmp_pfile = INIT_PATH_READER;
path_file_open(&tmp_pfile, optarg, true);
pfile_buf_add(&args->pfiles, tmp_pfile);
break;
case 't':
if(args->num_of_threads != 0) die("%s set twice", cmd);
args->num_of_threads = cmd_parse_arg_uint32_nonzero(cmd, optarg);
break;
case 'm': cmd_mem_args_set_memory(&args->memargs, optarg); break;
case 'n': cmd_mem_args_set_nkmers(&args->memargs, optarg); break;
case 'c': args->colour = cmd_parse_arg_uint32(cmd, optarg); break;
case '1':
case '2':
case 'i':
used = 1;
correct_aln_input_buf_add(inputs, task);
asyncio_task_parse(&inputs->data[inputs->len-1].files, c, optarg,
fq_offset, correct_cmd ? &tmp_path : NULL);
if(correct_cmd) inputs->data[inputs->len-1].out_base = tmp_path;
break;
case 'f': task.matedir = READPAIR_FR; used = 0; break;
case 'F': task.matedir = READPAIR_FF; used = 0; break;
case 'r': task.matedir = READPAIR_RF; used = 0; break;
case 'R': task.matedir = READPAIR_RR; used = 0; break;
case 'w': task.crt_params.one_way_gap_traverse = true; used = 0; break;
case 'W': task.crt_params.one_way_gap_traverse = false; used = 0; break;
case 'q': fq_offset = cmd_parse_arg_uint8(cmd, optarg); used = 0; break;
case 'Q': task.fq_cutoff = cmd_parse_arg_uint8(cmd, optarg); used = 0; break;
case 'H': task.hp_cutoff = cmd_parse_arg_uint8(cmd, optarg); used = 0; break;
case 'e': task.crt_params.use_end_check = true; used = 0; break;
case 'E': task.crt_params.use_end_check = false; used = 0; break;
case 'g': task.crt_params.ins_gap_min = cmd_parse_arg_uint32(cmd, optarg); used = 0; break;
case 'G': task.crt_params.ins_gap_max = cmd_parse_arg_uint32(cmd, optarg); used = 0; break;
case 'S': args->dump_seq_sizes = optarg; dump_seq_n++; break;
case 'M': args->dump_mp_sizes = optarg; dump_mp_n++; break;
case 'u': args->use_new_paths = true; break;
case 'C':
if(optarg == NULL || strcmp(optarg,"auto")) args->clean_threshold = -1;
else if(parse_entire_int(optarg,&tmp_thresh) && tmp_thresh >= -1) {
if(tmp_thresh != -1 && tmp_thresh < 2)
warn("Ignoring --clean %u (too small < 2)", tmp_thresh);
else if(tmp_thresh > 255)
warn("Ignoring --clean %u (too big > 255)", tmp_thresh);
else
args->clean_threshold = tmp_thresh;
}
else die("Bad argument for %s <auto|N> where N > 1", cmd);
args->clean_paths = (args->clean_threshold != 0);
break;
case 'X': gen_paths_print_contigs = true; break;
case 'Y': gen_paths_print_paths = true; break;
case 'Z': gen_paths_print_reads = true; break;
case ':': /* BADARG */
case '?': /* BADCH getopt_long has already printed error */
// cmd_print_usage(NULL);
die("`"CMD" thread -h` for help. Bad option: %s", argv[optind-1]);
default: abort();
}
}
if(args->num_of_threads == 0) args->num_of_threads = DEFAULT_NTHREADS;
// Check that optind+1 == argc
if(optind+1 > argc)
cmd_print_usage("Expected exactly one graph file");
else if(optind+1 < argc)
cmd_print_usage("Expected only one graph file. What is this: '%s'", argv[optind]);
char *graph_path = argv[optind];
status("Reading graph: %s", graph_path);
if(!used) cmd_print_usage("Ignored arguments after last --seq");
if(dump_seq_n > 1) die("Cannot specify --seq-gaps <out> more than once");
if(dump_mp_n > 1) die("Cannot specify --mp-gaps <out> more than once");
//
// Open graph graph file
//
GraphFileReader *gfile = &args->gfile;
graph_file_open(gfile, graph_path, true);
file_filter_update_intocol(&gfile->fltr, 0);
if(!correct_cmd && graph_file_usedcols(gfile) > 1)
die("Please specify a single colour e.g. %s:0", gfile->fltr.file_path.buff);
//
// Open path files
//
size_t path_max_usedcols = 0;
for(i = 0; i < args->pfiles.len; i++) {
// file_filter_update_intocol(&args->pfiles.data[i].fltr, 0);
if(!correct_cmd && path_file_usedcols(&args->pfiles.data[i]) > 1) {
die("Please specify a single colour e.g. %s:0",
args->pfiles.data[i].fltr.file_path.buff);
}
path_max_usedcols = MAX2(path_max_usedcols,
path_file_usedcols(&args->pfiles.data[i]));
}
args->path_max_usedcols = path_max_usedcols;
// Check for compatibility between graph files and path files
graphs_paths_compatible(gfile, 1, args->pfiles.data, args->pfiles.len);
// Check ins_gap_min < ins_gap_max
for(i = 0; i < inputs->len; i++)
{
CorrectAlnInput *t = &inputs->data[i];
t->files.ptr = t;
if(t->crt_params.ins_gap_min > t->crt_params.ins_gap_max) {
die("--min-ins %u is greater than --max-ins %u",
t->crt_params.ins_gap_min, t->crt_params.ins_gap_max);
}
correct_aln_input_print(&inputs->data[i]);
args->max_gap_limit = MAX2(args->max_gap_limit, t->crt_params.ins_gap_max);
}
}
