Update human_reference_genome.py

human_reference_genome.py

@@ -16,7 +16,6 @@
 
 from typing import List
 import datasets
-import gzip
 from Bio import SeqIO
 import regex as re
 
@@ -49,7 +48,7 @@ _URLS = {
     f"fasta": "https://ftp.ncbi.nlm.nih.gov/genomes/all/GCF/000/001/405/GCF_000001405.40_GRCh38.p14/GCF_000001405.40_GRCh38.p14_genomic.fna.gz"
 }
 
-_CHUNK_LENGTH = 6100
+_CHUNK_LENGTHS = [6000, 12000]
 _OVERLAP = 100
 _THRESHOLD_FILTER_N = 0.05
 
@@ -93,12 +92,33 @@ def continue_loop(split: str, chromosome: str) -> bool:
         return False
 
 
+class HumanReferenceGenomeConfig(datasets.BuilderConfig):
+    """BuilderConfig for The Human Reference Genome."""
+
+    def __init__(self, *args, chunk_length: int, **kwargs):
+        """BuilderConfig for The Pile.
+        Args:
+            chunk_length (:obj:`int`): Chunk length.
+            **kwargs: keyword arguments forwarded to super.
+        """
+        num_kbp = int(chunk_length/1000)
+        super().__init__(
+            *args,
+            name="+".join(f'{num_kbp}kbp'),
+            **kwargs,
+        )
+        self.chunk_length = chunk_length
+
+
 class HumanReferenceGenome(datasets.GeneratorBasedBuilder):
     """Human reference genome, filtered and split into chunks of consecutive
     nucleotides. The test set corresponds to chromosome 22, the validation set to
     chromosome 21 and all other chromosomes are used for training."""
 
     VERSION = datasets.Version("1.1.0")
+    BUILDER_CONFIG_CLASS = HumanReferenceGenomeConfig
+    BUILDER_CONFIGS = [HumanReferenceGenomeConfig(chunk_length=chunk_length) for chunk_length in _CHUNK_LENGTHS]
+    DEFAULT_CONFIG_NAME = "6kbp"
 
     def _info(self):
 
@@ -128,13 +148,13 @@ class HumanReferenceGenome(datasets.GeneratorBasedBuilder):
         downloaded_files = dl_manager.download_and_extract(urls_to_download)
 
         return [
-            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "train"}),
-            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "validation"}),
-            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "test"}),
+            datasets.SplitGenerator(name=datasets.Split.TRAIN, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "train", "chunk_length": self.config.chunk_length}),
+            datasets.SplitGenerator(name=datasets.Split.VALIDATION, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "validation", "chunk_length": self.config.chunk_length}),
+            datasets.SplitGenerator(name=datasets.Split.TEST, gen_kwargs={"filepath": downloaded_files['fasta'], "split": "test", "chunk_length": self.config.chunk_length}),
         ]
 
     # method parameters are unpacked from `gen_kwargs` as given in `_split_generators`
-    def _generate_examples(self, filepath, split):
+    def _generate_examples(self, filepath, split, chunk_length):
         with open(filepath, 'rt') as f:
             fasta_sequences = SeqIO.parse(f, 'fasta')
             # regex to filter lines of interest in the FASTA
@@ -160,14 +180,14 @@ class HumanReferenceGenome(datasets.GeneratorBasedBuilder):
                         seq_length = len(sequence)
 
                         # split into chunks
-                        num_chunks = (seq_length - 2 * _OVERLAP) // _CHUNK_LENGTH
-                        sequence = sequence[:(_CHUNK_LENGTH * num_chunks + 2 * _OVERLAP)]
+                        num_chunks = (seq_length - 2 * _OVERLAP) // chunk_length
+                        sequence = sequence[:(chunk_length * num_chunks + 2 * _OVERLAP)]
                         seq_length = len(sequence)
 
                         for i in range(num_chunks):
                             # get chunk
-                            start_pos = i * _CHUNK_LENGTH
-                            end_pos = min(seq_length, (i+1) * _CHUNK_LENGTH + 2 * _OVERLAP)
+                            start_pos = i * chunk_length
+                            end_pos = min(seq_length, (i+1) * chunk_length + 2 * _OVERLAP)
                             chunk_sequence = sequence[start_pos:end_pos]
 
                             # compute ratio of Ns
@@ -182,4 +202,3 @@ class HumanReferenceGenome(datasets.GeneratorBasedBuilder):
                                     'end_pos': end_pos
                                 }
                                 key += 1
-