Back

Fast-Part: Fast and Accurate Data Partitioning for Biological Sequence Analysis

Ahmed, S.; Emon, M. I.; Moumi, N. A.; ZHANG, L.

2025-07-08 bioinformatics
10.1101/2024.11.13.623463 bioRxiv
Show abstract

Developing effective machine learning models for classifications of biological sequences depends heavily on the quality of the training and test datasets split. Existing tools are either computationally expensive, unable to maintain the desired level of similarity between the training and test datasets, or unable to retain training-test ratio stratification. Here, we present Fast-Part, a fast and accurate sequence data partitioning tool that ensures strict homology separation between the training and test datasets and the best possible training: test stratification ratio, and at the same time, is computationally fast. Fast-Part demonstrates rapid and accurate partitioning performance across diverse protein sequence datasets and maintains strict partitioning compared to the existing tools. Fast-Part can handle massive datasets and maintain strict homology partitioning.

Matching journals

The top 3 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.