Multi-task adversarial autoencoder for functional genomic element generation with preserved biophysical properties.

@article{adamu2026multitask,
  title = {{Multi-task adversarial autoencoder for functional genomic element generation with preserved biophysical properties.}},
  author = {Shamsuddeen Adamu and Hitham Alhussian and Said Jadid Abdulkadir and Majdy Mohamed Eltayeb Eltahir and Sallam O F Khairy and Ibrahim Hayatu Hassan and Ibrahim Muhammad Kurah and Abubakar Mukhtar and Vijayakanthan Ganesalingam and Vaishali Ravi and Yahaya Saidu},
  journal = {BioData mining},
  year = {2026},
  month = aug,
  doi = {10.1186/s13040-026-00591-9},
  url = {https://europepmc.org/article/MED/42968110},
  note = {CC BY},
}

TY  - JOUR
TI  - Multi-task adversarial autoencoder for functional genomic element generation with preserved biophysical properties.
AU  - Shamsuddeen Adamu
AU  - Hitham Alhussian
AU  - Said Jadid Abdulkadir
AU  - Majdy Mohamed Eltayeb Eltahir
AU  - Sallam O F Khairy
AU  - Ibrahim Hayatu Hassan
AU  - Ibrahim Muhammad Kurah
AU  - Abubakar Mukhtar
AU  - Vijayakanthan Ganesalingam
AU  - Vaishali Ravi
AU  - Yahaya Saidu
PY  - 2026
DA  - 2026/08/03
JO  - BioData mining
DO  - 10.1186/s13040-026-00591-9
UR  - https://europepmc.org/article/MED/42968110
AB  - Generative modeling of genomic sequences presents a stringent test for deep learning, requiring the capture of long-range dependencies and functional constraints beyond local nucleotide statistics. Existing architectures frequently collapse to limited modes or reproduce shallow nucleotide distributions without encoding functional semantics. We introduce the Multi-Task Adversarial Autoencoder (MT-AAE), a hybrid generative framework that integrates adversarial regularization with auxiliary functional and biophysical objectives to enforce structured latent representations. Evaluated on an empirical human gene corpus, MT-AAE achieved a Train-on-Synthetic-Test-on-Real (TRTS) accuracy of 74.7%, compared with 41.0% for a standard GAN baseline. Stratified analysis further showed that functional discriminability increased to 89.3% when sequence lengths aligned with the model’s architectural window. Importantly, the learned representations exhibited emergent biological structure: synthetic sequences spontaneously preserved cis-regulatory syntax, including canonical TATA-box motifs recovered across 100% of generated promoter sequences without explicit rule encoding, though positional placement relative to the TSS was not statistically significant (KS p=0.90), and the high occurrence rate is partly attributable to the AT-rich composition of the generated sequences. Representation-level validation using frozen DNABERT-2 and DNABERT-S embeddings confirmed that the generated sequences retained functional information beyond shallow k-mer statistics. Cross-species evaluation on Mus musculus sequences further demonstrated species-specific learning consistent with known human–mouse regulatory divergence. The framework also mitigated mode collapse, maintaining near-uniform generation across functional classes (\documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}R_g ≈ 1.0\end{document}), including rare categories such as tRNAs ( < 2% of the dataset). These findings position MT-AAE as an effective framework for biologically constrained genomic sequence generation.  Supplementary Information: The online version contains supplementary material available at https://doi.org/10.1186/s13040-026-00591-9.
SN  - 1756-0381
ER  -