# paths
## dir
data_dir="data/10xgenomics_PBMC_5k_peaks"
results_dir="results/10xgenomics_PBMC_5k_peaks_classification_1"
## matrix files
file_mat_open=$data_dir/'peaks_rmsk_sampled_openchromatin_1kb_read_atac.mat'
file_mat_nucl=$data_dir/'peaks_rmsk_sampled_nucleosomes_1kb_fragment_center.mat'
file_mat_seq=$data_dir/'peaks_rmsk_sampled_sequences_1kb.mat'
## file with seeds
file_seed=$results_dir'/peaks_rmsk_sampled_seed.txt'

mkdir -p $results_dir
touch $file_seed

# EM param
n_iter='100'
n_shift='981'
n_core=24

# classify
for k in 10 20 30
do
	## results files
	file_prob=$results_dir/'peaks_rmsk_sampled_sequences_1kb_'$k'class_prob.mat4d'
	file_mod1=$results_dir/'peaks_rmsk_sampled_openchromatin_1kb_read_atac_'$k'class_model.mat'
	file_mod2=$results_dir/'peaks_rmsk_sampled_nucleosomes_1kb_fragment_center_'$k'class_model.mat'
	file_mod3=$results_dir/'peaks_rmsk_sampled_sequences_1kb_'$k'class_model.mat'	
	seed=$(< /dev/urandom tr -dc _A-Z-a-z-0-9 | head -c${1:-15};echo)
	echo "$file_prob $seed" >> $file_seed
	bin/EMSequence  --seq $file_mat_seq --class $k --shift $n_shift --flip --bgclass --iter $n_iter --seed $seed --thread $n_core > $file_prob
	bin/ProbToModel --read $file_mat_open --prob $file_prob --thread $n_core 1> $file_mod1
	bin/ProbToModel --read $file_mat_nucl --prob $file_prob --thread $n_core 1> $file_mod2
	bin/ProbToModel --seq  $file_mat_seq  --prob $file_prob --thread $n_core 1> $file_mod3
done