Biotext with SWeePtex: Bioinformatics Tricks to Perform Fast, Accurate, and Content-specific String Embedding
Authors:
- Diogo de J. S. MachadoDMDiogo de J. S. Machado
- Camilla R. De PierriCamilla R. De Pierri
- Antonio C. da Silva FilhoAFAntonio C. da Silva Filho
- Flávia de F. CostaFCFlávia de F. Costa
- Nelson A. de M. LemosNLNelson A. de M. Lemos
- Camila P. PericoCPCamila P. Perico
- Letícia G. C. SantosLSLetícia G. C. Santos
- Maricel G. KannMKMaricel G. Kann
- Fábio de O. PedrosaFPFábio de O. Pedrosa
- Roberto T. RaittzRRRoberto T. Raittz
Open Access
- 20 January 2026
- journal article
- Published by Sociedade Brasileira de Computação in Journal of the Brazilian Computer Society
- Vol. 32 (1) , 1155-1164
Abstract
The escalating demand for adaptable Artificial Intelligence (AI) systems presents a critical hurdle: generating efficient text embeddings tailored to specific problems. While Large Language Models (LLMs) excel in general contexts, they struggle in specialized domains due to their massive data requirements, opaque embedding strategies, and high computational costs. We introduce Biotext, featuring SWeePtex, a novel framework that adapts successful Bioinformatics techniques for text embedding. By converting text to the Biological Sequence-Like (BSL) format, our Python package enables the application of SWeeP, a tool originally developed for biological sequences, to create content-addressable vectors in natural language, employing the random projection paradigm. Using unsupervised machine learning, we validated this finding by analyzing data from 14,984 MEDLINE abstracts on the thioredoxin theme. Biotext, through SWeePtex, constructs a unified vector space for words and documents from scratch, capturing rich contextual relationships and offering scalable processing. Our usage example demonstrates that this Bioinformatics-inspired method effectively addresses key challenges in Natural Language Processing (NLP), providing interpretable, computationally efficient, and content-addressable linguistic representations for document exploration. Ultimately, Biotext demonstrates that bridging Bioinformatics and NLP yields powerful, efficient, and accessible text analysis tools that balance analytical power with interpretability, particularly valuable in specialized domains and resource-constrained environments. Biotext Python package is freely available at the PyPI repository.Keywords
This publication has 27 references indexed in Scilit:
- Forget ChatGPT: why researchers now run small AIs on their laptopsNature, 2024
- Reliability and Interpretability in Science and Deep LearningMinds and Machines, 2024
- Academic integrity and artificial intelligence: is ChatGPT hype, hero or heresy?Seminars in Nuclear Medicine, 2023
- Array programming with NumPyNature, 2020
- SWeeP: representing large biological sequences datasets in compact vectorsScientific Reports, 2020
- Unsupervised word embeddings capture latent knowledge from materials science literatureNature, 2019
- Biopython: freely available Python tools for computational molecular biology and bioinformaticsBioinformatics, 2009
- WEBSOM – Self-organizing maps of document collectionsNeurocomputing, 1998
- ICANN ’941994
- Extensions of Lipschitz mappings into a Hilbert space1984