static int
simulate_using_nussinov_scoring (const struct brot_args_info* brot_args,
SeqMatrix* sm, Scmf_Rna_Opt_data* data,
GFile* entropy_file)
{
int error = 0;
float** scores
= create_scoring_matrix (scmf_rna_opt_data_get_alphabet (data));
if (scores == NULL)
{
error = 1;
}
/* simulate */
if (!error)
{
scmf_rna_opt_data_set_scores (scores, data);
/* seqmatrix_set_func_calc_eeff_row (seqmatrix_calc_eeff_row_scmf, sm);*/
seqmatrix_set_func_calc_cell_energy (scmf_rna_opt_calc_nussinov, sm);
error = seqmatrix_simulate_scmf (brot_args->steps_arg,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
entropy_file,
sm,
data);
}
if (!error)
{
seqmatrix_set_transform_row (scmf_rna_opt_data_transform_row_2_base, sm);
error = seqmatrix_collate_is (0.99,
brot_args->steps_arg / 2,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
sm,
data);
/* error = seqmatrix_collate_mv (sm, sigma); */
}
scmf_rna_opt_data_set_scores (NULL, data);
XFREE_2D ((void**)scores);
return error;
}
int
brot_main(const char *cmdline)
{
struct brot_args_info brot_args;
SeqMatrix* sm = NULL;
int retval = 0;
Scmf_Rna_Opt_data* sim_data = NULL;
GFile* entropy_file = NULL;
/* command line parsing */
brot_cmdline_parser_init (&brot_args);
retval = brot_cmdline_parser_string (cmdline, &brot_args, get_progname());
if (retval == 0)
{
retval = brot_cmdline_parser_required (&brot_args, get_progname());
}
/* postprocess arguments */
if (retval == 0)
{
retval = brot_cmdline_parser_postprocess (&brot_args);
}
/* init simulation data */
if (retval == 0)
{
sim_data = SCMF_RNA_OPT_DATA_NEW_INIT(brot_args.inputs[1],
strlen (brot_args.inputs[1]),
RNA_ALPHABET,
strlen(RNA_ALPHABET)/2,
((-1) * ((logf (1 / 0.000001f)) / (strlen (brot_args.inputs[1])))));
if (sim_data == NULL)
{
retval = 1;
}
}
/* init matrix */
if (retval == 0)
{
sm = SEQMATRIX_NEW;
if (sm != NULL)
{
retval = SEQMATRIX_INIT (
alphabet_size (scmf_rna_opt_data_get_alphabet (sim_data)),
strlen (brot_args.inputs[1]),
sm);
/*seqmatrix_print_2_stdout (6, sm);*/
}
else
{
retval = 1;
}
}
/* fix certain sites in the matrix */
if (retval == 0)
{
retval = adopt_site_presettings (&brot_args,
scmf_rna_opt_data_get_alphabet (sim_data),
sm);
}
/* open entropy file if name given */
if (brot_args.entropy_output_given)
{
entropy_file = GFILE_OPEN (brot_args.entropy_output_arg,
strlen (brot_args.entropy_output_arg),
GFILE_VOID, "a");
if (entropy_file == NULL)
{
retval = 1;
}
else
{
if (gfile_printf (entropy_file, "# bROT settings:\n") < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# steps: %lu\n",
brot_args.steps_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# temp: %f\n",
brot_args.temp_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# beta-short: %f\n",
brot_args.beta_short_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# beta-long: %f\n",
brot_args.beta_long_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# speedup-threshold: %f\n",
brot_args.speedup_threshold_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# min-cool: %f\n",
brot_args.min_cool_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# scale-cool: %f\n",
brot_args.scale_cool_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# lambda: %f\n",
brot_args.lambda_arg) < 0)
{
retval = 1;
}
else if (gfile_printf (entropy_file, "# sm-entropy: %f\n",
brot_args.sm_entropy_arg) < 0)
{
retval = 1;
}
}
}
if (retval == 0)
{
if (brot_args.scoring_arg == scoring_arg_simpleNN)
{
/* special to NN usage: structure has to be of size >= 2 */
if (strlen (brot_args.inputs[1]) > 1)
{
retval = simulate_using_simplenn_scoring (&brot_args,
sm,
sim_data,
entropy_file);
}
else
{
THROW_ERROR_MSG (NN_2_SMALL_WARNING,
brot_args.inputs[1],
(unsigned long) strlen (brot_args.inputs[1]));
retval = 1;
}
}
else if (brot_args.scoring_arg == scoring_arg_nussinov)
{
retval = simulate_using_nussinov_scoring (&brot_args,
sm,
sim_data,
entropy_file);
}
else if (brot_args.scoring_arg == scoring_arg_NN)
{
/* special to NN usage: structure has to be of size >= 2 */
if (strlen (brot_args.inputs[1]) > 1)
{
retval = simulate_using_nn_scoring (&brot_args,
sm,
sim_data,
entropy_file);
}
else
{
THROW_ERROR_MSG (NN_2_SMALL_WARNING,
brot_args.inputs[1],
(unsigned long) strlen (brot_args.inputs[1]));
retval = 1;
}
}
}
/* close entropy file */
if (retval == 0)
{
retval = gfile_close (entropy_file);
}
else
{
gfile_close (entropy_file);
}
if (retval == 0)
{
/* seqmatrix_print_2_stdout (2, sm); */
mprintf ("%s\n", scmf_rna_opt_data_get_seq(sim_data));
}
/* finalise */
brot_cmdline_parser_free (&brot_args);
seqmatrix_delete (sm);
scmf_rna_opt_data_delete (sim_data);
if (retval == 0)
{
return EXIT_SUCCESS;
}
else
{
return EXIT_FAILURE;
}
}
static int
simulate_using_simplenn_scoring (struct brot_args_info* brot_args,
SeqMatrix* sm,
Scmf_Rna_Opt_data* data,
GFile* entropy_file)
{
int error = 0;
char** bp_allowed = NULL;
char bi, bj;
unsigned long i, j, k;
unsigned long alpha_size
= alphabet_size (scmf_rna_opt_data_get_alphabet(data));
unsigned long allowed_bp = 0;
NN_scores* scores =
NN_SCORES_NEW_INIT(0, scmf_rna_opt_data_get_alphabet (data));
if (scores == NULL)
{
error = 1;
}
/* prepare index of allowed base pairs */
if (!error)
{
bp_allowed = XMALLOC(alpha_size * sizeof (*bp_allowed));
if (bp_allowed == NULL)
{
error = 1;
}
}
if (!error)
{
allowed_bp = nn_scores_no_allowed_basepairs (scores);
/* we need 1 byte for each possible pair + 1byte for the NULL byte for
each letter in the alphabet */
bp_allowed[0] = XCALLOC (allowed_bp + alpha_size,
sizeof (**(bp_allowed)));
if (bp_allowed[0] == NULL)
{
error = 1;
}
}
if (!error)
{
i = 0;
k = 0;
while (i < alpha_size)
{
bp_allowed[i] = bp_allowed[0] + (k * sizeof (**(bp_allowed)));
for (j = 0; j < allowed_bp; j++)
{
nn_scores_get_allowed_basepair (j, &bi, &bj, scores);
if (i == (unsigned) bi)
{
bp_allowed[0][k] = bj + 1;
k++;
}
}
k++;
i++;
}
}
/* simulate */
if (!error)
{
scmf_rna_opt_data_set_scores (scores, data);
scmf_rna_opt_data_set_bp_allowed (bp_allowed, data);
seqmatrix_set_gas_constant (8.314472, sm);
seqmatrix_set_func_calc_cell_energy (scmf_rna_opt_calc_simplenn, sm);
/*scmf_rna_opt_data_init_negative_design_energies (data, sm);*/
seqmatrix_set_pre_col_iter_hook (
scmf_rna_opt_data_init_negative_design_energies_alt, sm);
error = seqmatrix_simulate_scmf (brot_args->steps_arg,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
entropy_file,
sm,
data);
}
/* collate */
if (!error)
{
seqmatrix_set_transform_row (scmf_rna_opt_data_transform_row_2_base, sm);
seqmatrix_set_fixed_site_hook (scmf_rna_opt_data_update_neg_design_energy,
sm);
error = seqmatrix_collate_is (0.99,
brot_args->steps_arg / 2,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
sm,
data);
/*error = seqmatrix_collate_mv (sm, data);*/
}
nn_scores_delete (scores);
scmf_rna_opt_data_set_scores (NULL, data);
scmf_rna_opt_data_set_bp_allowed (NULL, data);
XFREE (bp_allowed[0]);
XFREE (bp_allowed);
return error;
}
static int
simulate_using_nn_scoring (struct brot_args_info* brot_args,
SeqMatrix* sm,
Scmf_Rna_Opt_data* data,
GFile* entropy_file)
{
int error = 0;
char** bp_allowed = NULL;
char bi, bj;
unsigned long i, j, k;
unsigned long alpha_size
= alphabet_size (scmf_rna_opt_data_get_alphabet(data));
unsigned long allowed_bp = 0;
NN_scores* scores =
NN_SCORES_NEW_INIT(50.0f, scmf_rna_opt_data_get_alphabet (data));
if (scores == NULL)
{
error = 1;
}
/* prepare index of allowed base pairs */
if (!error)
{
/* "randomise" scoring function */
mfprintf (stdout, "Using seed: %ld\n", brot_args->seed_arg);
nn_scores_add_thermal_noise (alpha_size, brot_args->seed_arg, scores);
bp_allowed = XMALLOC(alpha_size * sizeof (*bp_allowed));
if (bp_allowed == NULL)
{
error = 1;
}
}
if (!error)
{
allowed_bp = nn_scores_no_allowed_basepairs (scores);
/* we need 1 byte for each possible pair + 1byte for the NULL byte for
each letter in the alphabet */
bp_allowed[0] = XCALLOC (allowed_bp + alpha_size,
sizeof (**(bp_allowed)));
if (bp_allowed[0] == NULL)
{
error = 1;
}
}
if (!error)
{
i = 0;
k = 0;
while (i < alpha_size)
{
bp_allowed[i] = bp_allowed[0] + (k * sizeof (**(bp_allowed)));
for (j = 0; j < allowed_bp; j++)
{
nn_scores_get_allowed_basepair (j, &bi, &bj, scores);
if (i == (unsigned) bi)
{
bp_allowed[0][k] = bj + 1;
k++;
}
}
k++;
i++;
}
}
/* decompose secondary structure */
if (!error)
{
error = scmf_rna_opt_data_secstruct_init (data);
}
mfprintf (stdout, "TREATMENT of fixed sites: If both sites of a pair are "
"fixed, delete from list? Verbose info!!!\n");
/* set our special function for calc. cols.: Iterate over sec.struct., not
sequence matrix! */
if (!error)
{
scmf_rna_opt_data_set_scores (scores, data);
scmf_rna_opt_data_set_bp_allowed (bp_allowed, data);
scmf_rna_opt_data_set_scales (brot_args->negative_design_scaling_arg,
brot_args->heterogenity_term_scaling_arg,
data);
scmf_rna_opt_data_set_het_window (brot_args->window_size_arg, data);
seqmatrix_set_func_calc_eeff_col (scmf_rna_opt_calc_col_nn, sm);
seqmatrix_set_gas_constant (8.314472, sm);
error = seqmatrix_simulate_scmf (brot_args->steps_arg,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
entropy_file,
sm,
data);
}
/* collate */
if (!error)
{
/* seqmatrix_print_2_stdout (2, sm); */
seqmatrix_set_transform_row (scmf_rna_opt_data_transform_row_2_base, sm);
error = seqmatrix_collate_is (0.99,
brot_args->steps_arg / 2,
brot_args->temp_arg,
brot_args->beta_long_arg,
brot_args->beta_short_arg,
brot_args->speedup_threshold_arg,
brot_args->min_cool_arg,
brot_args->scale_cool_arg,
brot_args->lambda_arg,
brot_args->sm_entropy_arg,
sm,
data);
/*error = seqmatrix_collate_mv (sm, data);*/
}
/* first: iterate scmf on secstruct, not sm! */
nn_scores_delete (scores);
scmf_rna_opt_data_set_scores (NULL, data);
scmf_rna_opt_data_set_bp_allowed (NULL, data);
XFREE (bp_allowed[0]);
XFREE (bp_allowed);
return error;
}
int
brot_main(const char *cmdline)
{
struct brot_args_info brot_args;
SeqMatrix* sm = NULL;
int retval = 0;
Scmf_Rna_Opt_data* sim_data = NULL;
GFile* entropy_file = NULL;
GFile* simulation_file = NULL;
/* command line parsing */
brot_cmdline_parser_init (&brot_args);
retval = brot_cmdline_parser_string (cmdline, &brot_args, get_progname());
if (retval == 0)
{
retval = brot_cmdline_parser_required (&brot_args, get_progname());
}
/* postprocess arguments */
if (retval == 0)
{
retval = brot_cmdline_parser_postprocess (&brot_args, cmdline);
}
/* init simulation data */
if (retval == 0)
{
sim_data = SCMF_RNA_OPT_DATA_NEW_INIT(brot_args.inputs[1],
strlen (brot_args.inputs[1]),
RNA_ALPHABET,
strlen(RNA_ALPHABET)/2,
((-1) * ((logf (1 / 0.000001f)) / (strlen (brot_args.inputs[1])))),
brot_args.file_given);
if (sim_data == NULL)
{
retval = 1;
}
}
/* init matrix */
if (retval == 0)
{
sm = SEQMATRIX_NEW;
if (sm != NULL)
{
retval = SEQMATRIX_INIT (
alphabet_size (scmf_rna_opt_data_get_alphabet (sim_data)),
scmf_rna_opt_data_get_rna_size(sim_data),
/* strlen (brot_args.inputs[1]), */
sm);
/*seqmatrix_print_2_stdout (6, sm);*/
}
else
{
retval = 1;
}
}
/* fix certain sites in the matrix */
if (retval == 0)
{
retval = adopt_site_presettings (&brot_args,
scmf_rna_opt_data_get_alphabet (sim_data),
sim_data,
sm);
}
/* open entropy file if name given */
if (retval == 0)
{
print_verbose ("# Entropy file (-p): ");
if (brot_args.entropy_output_given)
{
print_verbose ("%s", brot_args.entropy_output_arg);
entropy_file = GFILE_OPEN (brot_args.entropy_output_arg,
strlen (brot_args.entropy_output_arg),
GFILE_VOID, "a");
if (entropy_file == NULL)
{
retval = 1;
}
else
{
retval = brot_settings_2_file (entropy_file, cmdline);
}
}
/* open simulation/ matrix file if name given */
print_verbose ("\n# Simulation/ Matrix file (-m): ");
if (brot_args.simulation_output_given)
{
print_verbose ("%s", brot_args.simulation_output_arg);
simulation_file = GFILE_OPEN (brot_args.simulation_output_arg,
strlen (brot_args.simulation_output_arg),
GFILE_VOID, "a");
if (simulation_file == NULL)
{
retval = 1;
}
else
{
retval = brot_settings_2_file (simulation_file, cmdline);
if (retval == 0)
{
if (gfile_printf (simulation_file, "START\n") < 0)
{
retval = 1;
}
}
}
}
}
if (retval == 0)
{
print_verbose ("\n# Scoring scheme (-c): ");
if (brot_args.scoring_arg == scoring_arg_simpleNN)
{
print_verbose ("simpleNN\n");
/* special to NN usage: structure has to be of size >= 2 */
if (strlen (brot_args.inputs[1]) > 1)
{
retval = simulate_using_simplenn_scoring (&brot_args,
sm,
sim_data,
entropy_file,
simulation_file);
}
else
{
THROW_ERROR_MSG (NN_2_SMALL_WARNING,
brot_args.inputs[1],
(unsigned long) strlen (brot_args.inputs[1]));
retval = 1;
}
}
else if (brot_args.scoring_arg == scoring_arg_nussinov)
{
print_verbose ("nussinov\n");
retval = simulate_using_nussinov_scoring (&brot_args,
sm,
sim_data,
entropy_file,
simulation_file);
}
else if (brot_args.scoring_arg == scoring_arg_NN)
{
print_verbose ("NN\n");
/* special to NN usage: structure has to be of size >= 2 */
if (strlen (brot_args.inputs[1]) > 1)
{
retval = simulate_using_nn_scoring (&brot_args,
sm,
sim_data,
entropy_file,
simulation_file);
}
else
{
THROW_ERROR_MSG (NN_2_SMALL_WARNING,
brot_args.inputs[1],
(unsigned long) strlen (brot_args.inputs[1]));
retval = 1;
}
}
}
/* close files */
if (retval == 0)
{
retval = gfile_close (entropy_file);
}
else
{
gfile_close (entropy_file);
}
if (simulation_file != NULL)
{
if (gfile_printf (simulation_file, "END\n") < 0)
{
retval = 1;
}
}
if (retval == 0)
{
retval = gfile_close (simulation_file);
}
else
{
gfile_close (simulation_file);
}
if (retval == 0)
{
/*seqmatrix_print_2_stdout (2, sm);*/
mprintf ("%s\n", scmf_rna_opt_data_get_seq(sim_data));
}
/* finalise */
brot_cmdline_parser_free (&brot_args);
seqmatrix_delete (sm);
scmf_rna_opt_data_delete (sim_data);
if (retval == 0)
{
return EXIT_SUCCESS;
}
else
{
return EXIT_FAILURE;
}
}
