Boosting t-SNE Efficiency for Sequencing Data: Insights from Kernel Selection

TL;DR

This paper evaluates nine kernel functions for t-SNE in biological sequence analysis, finding that cosine similarity offers superior efficiency and data preservation over traditional kernels across multiple datasets and embedding methods.

Contribution

It provides a comprehensive comparison of kernel functions for t-SNE on sequencing data, highlighting the effectiveness of cosine similarity for visualization and downstream tasks.

Findings

Cosine similarity kernel outperforms Gaussian and isolation kernels in visualization quality.

Cosine kernel achieves better runtime efficiency and pairwise distance preservation.

Kernel choice significantly affects downstream classification and clustering performance.

Abstract

Dimensionality reduction techniques are essential for visualizing and analyzing high-dimensional biological sequencing data. t-distributed Stochastic Neighbor Embedding (t-SNE) is widely used for this purpose, traditionally employing the Gaussian kernel to compute pairwise similarities. However, the Gaussian kernel's lack of data-dependence and computational overhead limit its scalability and effectiveness for categorical biological sequences. Recent work proposed the isolation kernel as an alternative, yet it may not optimally capture sequence similarities. In this study, we comprehensively evaluate nine different kernel functions for t-SNE applied to molecular sequences, using three embedding methods: One-Hot Encoding, Spike2Vec, and minimizers. Through both subjective visualization and objective metrics (including neighborhood preservation scores), we demonstrate that the cosine similarity kernel in general outperforms other kernels, including Gaussian and isolation kernels, achieving superior runtime efficiency and better preservation of pairwise distances in low-dimensional space. We further validate our findings through extensive classification and clustering experiments across six diverse biological datasets (Spike7k, Host, ShortRead, Rabies, Genome, and Breast Cancer), employing multiple machine learning algorithms and evaluation metrics. Our results show that kernel selection significantly impacts not only visualization quality but also downstream analytical tasks, with the cosine similarity kernel providing the most robust performance across different data types and embedding strategies, making it particularly suitable for large-scale biological sequence analysis.