Brhanu Fentaw Znabu
Ph.D. student in Biomedical Engineering at the University of Nebraska–Lincoln,
advised by Nicole R. Sexton at the
Nebraska Center for Virology.
I build and evaluate genomic foundation models for biological sequence analysis.
My current work focuses on domain-adaptive pre-training of large language models (DNABERT-2) for
viral host-range prediction and epidemic emergence forecasting, with an emphasis on
rigorous leakage-aware evaluation and model interpretability.
News
Apr 2026
Nominated by the University of Nebraska–Lincoln for the Google PhD Fellowship 2026 in AI for Health.
Apr 2026
Served as a Poster Judge at the School of Biological Sciences Undergraduate Research Symposium, UNL.
Mar 2026
Competed in the UNL Engineering Pitch Competition with Traversa, a startup building AI-driven infectious disease risk assessment for travelers.
Feb 2026
Poster presentation at the Annual Engineering Symposium, UNL: “Deciphering the host-switching grammar of flaviviruses using foundation model embeddings.”
Oct 2025
Poster presentation at the NCV Annual Virology Symposium: “Cracking the viral code: codon usage and CpG patterns predict host specificity.”
Apr 2025
Paper on MoSeq-based 3D behavioral profiling in diabetic neuropathy published in Scientific Reports.
2025
Submitted manuscript on leakage-aware evaluation of foundation model embeddings for flavivirus host-range prediction to PLOS Computational Biology.
Aug 2024
Joined the Sexton Lab as a Ph.D. student at the Nebraska Center for Virology, UNL.
Publications
In Preparation
A genomic foundation model for predicting arbovirus epidemic emergence across RNA virus families
Znabu BF, Sexton NR.
Target: Bioinformatics (Oxford)
ArboFM: domain-adapted DNABERT-2 via continued MLM pre-training on 120K+ arbovirus genome windows (9,299 genomes, 362 species, 6 families).
Predicts epidemic emergence with AP = 0.978 (Flaviviridae). Retrospective temporal validation detects Zika, chikungunya,
and West Nile epidemic lineages before documented emergence.
In Preparation
Deciphering the host-switching grammar of flaviviruses using foundation model embeddings: a leakage-aware evaluation framework
Znabu BF, Sexton NR.
Target: PLOS Computational Biology
DNABERT-2 embeddings on 3,031 flavivirus genomes. Quantifies 15-percentage-point performance inflation from phylogenetic leakage.
Identifies UpA-containing motifs linked to the OAS/RNaseL innate immunity pathway via genome-localized attribution.
In Preparation
Cross-family comparison of host-switching grammar in arthropod-borne viruses reveals convergent evolution of family-specific immune evasion strategies
Znabu BF, Sexton NR.
Target: Virus Evolution (Oxford)
Cross-family transfer learning across 4 arbovirus families (6,052 genomes, 124,883 windows).
Demonstrates host-switching grammar is predominantly family-specific (Jaccard ≈ 0), supporting convergent evolution.
In Preparation
Genome-scale classification of flavivirus host range using composition and codon-usage signatures
Znabu BF, Sexton NR.
Target: Journal of Virology
Species-stratified ML classifiers on 1,285 curated flavivirus genomes with 97 compositional features.
Achieves PR-AUC = 1.000 for ISFV vs. dual-host classification. CpG suppression identified as the primary discriminative feature.
Published 2025
MoSeq based 3D behavioral profiling uncovers neuropathic behavior changes in diabetic mouse model
Ashiquzzaman A*, Lee E*, Znabu BF*, Sakib AN, Chung G, Kim SS, Kim YR, Kwon H-S, Chung E.
Scientific Reports 15, 15114 (2025)
Preprint
Interpretable deep learning-based multi-omics integration for prognosis in hepatocellular carcinoma
Znabu BF, Atif Z.
bioRxiv (2026)
Attention-based multi-branch deep learning framework integrating multi-omics data for interpretable survival prediction in hepatocellular carcinoma.
About
My research sits at the intersection of machine learning and genomics. I develop foundation models and rigorous evaluation frameworks
for biological sequence analysis, with applications in viral emergence prediction, host-range classification, and model interpretability.
I work primarily with PyTorch, Hugging Face Transformers, and the SciPy/PyData ecosystem on HPC GPU clusters.
Prior to my PhD, I earned an M.Sc. in Biomedical Science and Engineering from
Gwangju Institute of Science and Technology (GIST), South Korea,
where I applied unsupervised machine learning (autoregressive models, Hidden Markov models) to 3D behavioral phenotyping
in diabetic neuropathy mouse models. I also hold a B.Sc. in Biomedical Engineering from
Jimma University, Ethiopia, graduating ranked 4th of 180.
When I'm not at the computer, you'll find me exploring coffee shops ☕ or at the gym 🏋.
Awards & Grants
2020
Korean Government Full Scholarship for Master's Study, GIST
2020
Research Grant for Colostomy Device Development, Hawassa University
2018
Best B.Sc. Thesis Award, Ethiopian Science, Technology & Innovation
2018
Graduated with Distinction, Ranked 4th of 180, Jimma University