Dataset Viewer
Auto-converted to Parquet Duplicate
task
stringclasses
1 value
seq
stringlengths
8.19k
131k
label
int64
0
2
sample_name
stringclasses
146 values
hap
stringclasses
1 value
chr
stringclasses
1 value
start_pos
int64
1
151M
end_pos
int64
8.19k
151M
Human_classify
AGGGGTACGGGGTTTGTGGGTGACGGTGACCGTTGCCTGGGGCCGAGGGGAATGGGGGCGTCTGTGTGAGGGGTACAGGGTTTGCGGGTGACGGTGACCGTTGCCTGGGGCCGAGGGGAATGGGGGCGTCTGTGTGAGGGGTACGGGGTTTGTGGGTGACGGTGACCGTTGCCTGGGGCTGAGGGGAATGGGGCGTCTGTGTGAGGGGTACGGGGTTTGCGGGTGACGGTGACCGTTGCCTGGGGCCGAGGGGAATGGGGCGTCTGTGTGAGGGGTACGGGGTTTGCTTCTGAAGCGATGAACCATTCTGAAACCAGAAG...
0
HG00344
hap1
chr9
149,487,617
149,618,688
Human_classify
"CTAACCCTAACCCTAACCCTAACCCTAACCCTAACCTAACCCTAACCCTAACCCCTAACCCTAACCCTAACCCTAACCCTAACCCTAACCCTAACCCTA(...TRUNCATED)
0
HG00344
hap1
chr9
1
131,072
Human_classify
"ATCAAATATTGTCATAAAATGTAATTTTGATGGTTGCATAGATTCCTTTTTATGGATGTAATTTATGTGCCTTATTTTAATATTCTTTTTATTGGTAGA(...TRUNCATED)
0
HG00344
hap1
chr9
327,681
458,752
Human_classify
"GCTGGCATGTCTCTGGTCAGGGAGGGGTTTATCCTATGGTTGGAATGTTTCTGGTTGGACATGTCATTTGTGGTTTATGGTCATGCTGACCTTAGCCAT(...TRUNCATED)
0
HG00344
hap1
chr9
39,714,817
39,845,888
Human_classify
"TGAGTCTGAGGAGCCAATGTTTCAGCTGAGTCTGAAGACAGAAGACACCAATGTCCAAGCTCAAACCACCAGGCAGGAGGAGTTCTTTCTTATTTGCAC(...TRUNCATED)
0
HG00344
hap1
chr9
39,649,281
39,780,352
Human_classify
"CCCGTCTCCACTAAAAAATACCAAAAAATTAGCCAGGCGTGGTGGCGGGCACCTGTAATCCTAGCTACTCGGGAGGCTGAGGCAGAAGAATGGCGTGAA(...TRUNCATED)
0
HG00344
hap1
chr9
39,518,209
39,649,280
Human_classify
"GATCTTCAAGGGCTGGTCCCCGGCCCTAGACACAGGGAACCCATGAAGAAGCCCGGCAGAATGTTATTCGTCAGCCTCTTCTGGTTCCCAAACAAGTCC(...TRUNCATED)
0
HG00344
hap1
chr9
149,684,225
149,815,296
Human_classify
"TTCCTATGACCCAAAGGAGAGTAGCCATACTTCTTATCACTTATTTCTTCAAGTGCACACACATCATTTAAACCTAATTTCCTACCATTTCTTCCTAGT(...TRUNCATED)
0
HG00344
hap1
chr9
28,114,945
28,246,016
Human_classify
"GTGCCCTTGCACTCCAGCCTGGGCGACAGAGCAAGACTCCATCTCAAAAAAGAACAATATAATAAAAAATAAATCCCAGAATTATGTAAAGTGCTGATT(...TRUNCATED)
0
HG00344
hap1
chr9
38,928,385
39,059,456
Human_classify
"GACAGCCCAGAGACGCTGCCTTCCCACAAGGCGACCTGACACTGTGTCTGAACACAGGTGACAAGAGACACTGTGCCCGAGGGGCCAGGACACGTTGGA(...TRUNCATED)
0
HG00344
hap1
chr9
150,011,905
150,142,976
End of preview. Expand in Data Studio

Sumary

This dataset provides a benchmark for evaluating the model's ability to leverage richer genetic information from longer sequences to achieve more accurate inference. Using data from the Human Pangenome Reference Consortium (BioProject ID: PRJNA730823), we designed a population classification task focusing on African, East Asian, and European population groups. From samples' VCF file and the reference genome sequence, we generated sample pseudo-sequences. Based on variant site information recorded in the VCF file, we extracted a variant-dense region from chromosome 9. We used three sequence lengths: 8,192 bp (8K), 32,768 bp (32K), and 131,072 bp (128K). An XGBoost classifier was employed to perform classification predictions for individual sequences.

Usage

from datasets import load_dataset

#Download the whole dataset
dataset = load_dataset("BGI-HangzhouAI/Benchmark_Dataset-Human_population_classification")

#Download a specific task
task_name = "Human_population_classification_8192"
dataset = load_dataset(
    "BGI-HangzhouAI/Benchmark_Dataset-Human_population_classification",
    data_files = {
        "train": f"{task_name}/train.jsonl",
        "test": f"{task_name}/test.jsonl",
        "eval": f"{task_name}/eval.jsonl",
    }
)

Benchmark tasks

Task task_name Input fields # Train Seqs # Validation Seqs # Test Seqs
Human_population_classification 8k Human_population_classification_8192 {seq, label} 23,172 2,906 2,916
Human_population_classification 32k Human_population_classification_32768 {seq, label} 23,207 2,913 2,925
Human_population_classification 128k Human_population_classification_131072 {seq, label} 23,623 2,830 2,957
Population Sample_counts Label
EUR-European 30 0
AFR-African 69 1
EAS-East Asian 50 2

Data processing

1. Pseudo-sequence Generation:

For each sample, pseudo-sequences (including hap1 and hap2) were generated from its VCF file and the reference genome sequence using bcftools.

2. VCF Variant Region Statistics:

Using the sample VCF files, sliding windows of three different lengths (8K, 32K, 128K) were applied from the start of chromosome 9 of the reference genome. The overlap between consecutive windows was half the window length. The number of variants within each window was counted. Windows were then ranked in descending order based on this variant count to identify variant-dense genomic coordinates. Chromosome 9 were randomly selected, other autosomes could also be used.

3. Centromere Removal:

Centromeric regions, which are repetitive and non-coding and thus unsuitable for variant or classification tasks, were filtered out according to a BED file. This resulted in a final mapping of genomic windows to their corresponding variant counts.

4. Data Selection:

The samples for each label were split into training, validation and test sets in a 8:1:1 ratio. Based on the previously obtained window-variant count mapping, the hap1 pseudo-sequences for chromosome 9 of each sample were segmented. Regions were selected starting from the highest variant count downwards, while ensuring a roughly balanced number of sequences for each label.

5. Final format

Datasets are saved in JSONL format. Each file contains:

  • "seq" — the DNA sequence string (A/C/G/T, uppercase)
  • "label" — ternary class indicator (0 = CEU, 1 = AFR, 2 = EAS)
Downloads last month
28