JMIR Publications

ABSTRACT

Background:

DNA aptamers are short, single-stranded oligonucleotides that bind molecular targets with high affinity and specificity. Traditionally, aptamers are identified through SELEX (Systematic Evolution of Ligands by EXponential enrichment), an iterative process that is time-consuming and requires specialized equipment. To simplify and accelerate aptamer discovery, particularly across multiple mo-lecular targets, computational approaches have gained momentum.

Objective:

In this work, we present a sequence-only framework for classifying DNA ap-tamers by target type using curated datasets of experimentally validated sequences.

Methods:

Each sequence is represented using character-level TG-IDF of n-gram motifs. An ensemble of XGBoost classifiers is trained on these representations to distinguish between aptamers targeting cellular and steroidal molecules.

Results:

The model was further validated on a set of 13 newly selected aptamers for ster-oid aldosterone using experimental SELEX, achieving consistent predictive per-formance, confirming its ability to generalize new experimental data.

Conclusions:

These results demonstrate that informative target-specific patterns can be extract-ed directly from DNA sequences, establishing a scalable, data-driven comple-ment to SELEX for rapid pre-screening of candidate aptamers. Clinical Trial: None