datasets
Training and evaluation data, with the modality, task and licence stated up front. Listed live from the Hugging Face Hub.
Evolutionary
Evolutionary MSA Data
This repository contains precomputed evolutionary sequence-alignment data in an archive format that is practical to host and download from the Hub. The original file paths are preserved inside the tar shard, while metadata.csv gives a searchable index of every file.
The dataset is meant for workflows that need ready-to-use MSA/cache files without rebuilding them from sequence databases.
Contents
Component
Files
Size
msa_cache/
134,898… See the full description on the dataset page: https://huggingface.co/datasets/LiteFold/Evolutionary.evolutionary-constraint
Evolutionary constraint prediction
Reproducing this dataset
import pandas as pd
CENTER_POSITION = 255
NUCLEOTIDES = list("ACGT")
df = pd.read_csv("hf://datasets/kuleshov-group/cross-species-single-nucleotide-annotation/Evolutionary_constraint/valid.tsv", sep="\t")
df = df.rename(columns={"sequences": "seq"})
df["chrom_pos"] = df.pos
df["pos"] = CENTER_POSITION
df["ref"] = df.seq.str[CENTER_POSITION]
def subsample(df, n, seed):
return (… See the full description on the dataset page: https://huggingface.co/datasets/plantcad/evolutionary-constraint.