| 66 | return np.array(feature_matrix) |
| 67 | |
| 68 | def prepare_input(data_set, params): |
| 69 | if params['encode'] == 'one-hot': |
| 70 | seq_matrix = np.array(data_set['Sequence'].apply(one_hot_encode).tolist()) # (number of sequences, length of sequences, nucleotides) |
| 71 | elif params['encode'] == 'k-mer': |
| 72 | seq_matrix = np.array(data_set['Sequence'].apply(kmer_features, k=3).tolist()) # (number of sequences, 1, 4^k) |
| 73 | else: |
| 74 | raise Exception ('wrong encoding method') |
| 75 | |
| 76 | Y_dev = data_set.Dev_log2_enrichment |
| 77 | Y_hk = data_set.Hk_log2_enrichment |
| 78 | Y = [Y_dev, Y_hk] |
| 79 | |
| 80 | return seq_matrix, Y |
| 81 | |
| 82 | def DeepSTARR(params): |
| 83 | if params['encode'] == 'one-hot': |