Dr. Taeho Jo delivered an invited online lecture on applying AI to professional practice at the Korea Human Resource Development Institute for Health & Welfare on March 23, 2026. The session was attended by healthcare professionals and educators.

Vibe Coding with Claude Code (Hanbit Media, 2025) reached #1 on the Kyobo Bookstore Tech & Computer Monthly Bestseller chart for February 2026. Kyobo is South Korea's largest bookstore chain, and its monthly rankings are widely regarded as the national standard for book sales.

The book originated from Dr. Jo's experience developing AlphaGenome MCP, a Model Context Protocol server for genomic analysis built with Claude Code. The process of building a research tool through AI-assisted coding led to a broader guide on the subject, covering the full workflow from environment setup to web application deployment. Its reception reflects growing public interest in how AI tools are reshaping software development practices.

A companion site, vibeindex.ai, provides readers with curated AI tool recommendations and supplementary resources.

"DuAL-Net: A Dual-Network Approach for Alzheimer's Disease Risk Prediction Using APOE-Centered Regional WGS Data"

Authors: Eun Hye Lee, Taeho Jo*

Published in: Computational and Structural Biotechnology Journal (2026)

This study presents DuAL-Net (Dual Approach Local-global Network), a hybrid deep learning framework for Alzheimer's disease risk prediction using whole-genome sequencing (WGS) data. The framework integrates local genomic patterns with global biological knowledge by combining TabNet and Random Forest models. Analyzing 1,566 participants from ADNI and ADSP, with validation in 5,570 individuals from ADSP ADC, DuAL-Net addresses the high dimensionality of WGS data through a novel dual-network approach that examines both non-overlapping genomic windows and annotation-guided SNP groupings within the APOE region. The study identifies key genetic variants contributing to AD prediction and demonstrates that annotation-guided scoring improves identification of disease-relevant variants.

Dr. Donghyun Lee joined the lab as a visiting scholar. With an MD from Ulsan University College of Medicine and clinical experience at Asan Medical Center and Hallym University Gangnam Sacred Heart Hospital, he will focus on studying AI applications in Alzheimer's disease from a clinical perspective.

Leonard Lee joined the lab as a research intern. With a Bachelor's degree in Computer Science from Purdue University and currently pursuing a Master's in Artificial Intelligence at Johns Hopkins University, he will focus on studying Alzheimer's disease data analytics and artificial intelligence.

Dr. Jo published Vibe Coding with Claude Code through Hanbit Media. The book grew out of his experience developing AlphaGenome MCP, a Model Context Protocol server for genomic analysis built entirely with AI-assisted coding. That research tool development process inspired a broader guide covering the full workflow from environment setup to web application deployment.

Upon release, the book entered the Top 10 IT bestseller list in Korea and has received endorsements from healthcare professionals, students, and industry leaders.

Dr. Taeho Jo was invited to present a special online lecture on AI in healthcare at Seoul National University Bundang Hospital on November 19, 2025. The virtual session brought together data scientists and healthcare professionals to explore practical principles for implementing AI in clinical settings.

The presentation covered key topics including:

• The transformation process from patient data to actionable insights
• Real-world applications of AI empowering users in scientific research
• Case studies from Indiana University School of Medicine's AI applications in Alzheimer's disease research
• Practical strategies for personalized AI adoption and immediate implementation opportunities

The lecture provided attendees with a comprehensive framework for understanding how AI can be effectively integrated into medical practice, emphasizing both theoretical foundations and practical applications that healthcare professionals can begin implementing immediately.

Dr. Taeho Jo was invited as a keynote speaker at the AI IN MEDICINE 2025: AI Shock - How Healthcare Is Changing Now symposium hosted by Seoul Asan Medical Center. The symposium was held in Seoul, South Korea on October 30, 2025, with Dr. Jo's keynote presentation taking place from 14:30-15:30 during Session 2.

The presentation, titled "Medical Biotechnology: 3 Principles for Successful AI Implementation", introduced key strategies for effectively integrating AI technologies in medical and biotechnology applications. Dr. Jo's invitation as a keynote speaker at this prestigious symposium reflects the international recognition of his research in AI applications for Alzheimer's disease and genomic analysis at the Indiana Alzheimer's Disease Research Center.

"Uncertainty-Aware Genomic Classification of Alzheimer's Disease: A Transformer-Based Ensemble Approach with Monte Carlo Dropout"

Authors: Taeho Jo*, Eun Hye Lee

This study presents TrUE-Net, an innovative deep learning framework that addresses a critical challenge in genomic-based Alzheimer's disease prediction: the lack of confidence estimates in model predictions. By combining transformer and random forest models with Monte Carlo Dropout for uncertainty quantification, the framework analyzed whole-genome sequencing data from 1,050 individuals (607 AD, 443 controls) from the Alzheimer's Disease Neuroimaging Initiative (ADNI) cohort. The model demonstrated the ability to distinguish between "certain" and "uncertain" predictions, achieving 72.9% accuracy (F1=0.821) on high-confidence predictions while maintaining transparency about cases where predictions are less reliable.

Key Innovation: TrUE-Net introduces uncertainty quantification to genomic AD classification, enabling clinicians and researchers to identify which predictions are most trustworthy—a crucial advancement for translating AI predictions into clinical decision-making where understanding model confidence is essential for patient care. 

"AlphaGenome MCP: A Model Context Protocol Server for Conversational Genomics"

We introduce AlphaGenome MCP, a Model Context Protocol server that provides natural language access to AlphaGenome's gene expression prediction capabilities through Claude Desktop and Claude Code. This tool eliminates the need for coding expertise, enabling researchers to perform complex genomic analyses, including in silico mutagenesis, tissue-specific expression prediction, and variant effect assessment, through simple conversational prompts. Demonstrated on ABI3 gene analysis, the server successfully replicated key findings from our Alzheimer's disease genomics research within minutes of interaction.

Key Innovation: AlphaGenome MCP bridges the gap between sophisticated AI genomics tools and non-computational researchers by translating natural language requests into API calls, allowing instant exploratory analysis without programming requirements while maintaining full access to AlphaGenome's pre-trained deep learning models.  [LINK]

"Longitudinal plasma proteomics: relation to incident Alzheimer disease dementia and biomarkers"

First author: Dr. Eun Hye Lee (Postdoctoral Fellow, Dr. Jo's Medical AI Lab)

Authors: Eun Hye Lee, Yen-Ning Huang, Tamina Park, Shiwei Liu, Nicholas Adzibolosu, Soumilee Chaudhuri, Changgee Chang, Paula J. Bice, Jeffrey L. Dage, Jared R. Brosch, Sujuan Gao, Liana G. Apostolova, Donna M. Wilcock, Shannon L. Risacher, Andrew J. Saykin*, Taeho Jo*, and Kwangsik Nho*

This study presents a comprehensive analysis of longitudinal plasma proteomics data from 347 participants at the Indiana Alzheimer's Disease Research Center to identify molecular signatures associated with Alzheimer's disease progression. The research identified two proteins (ACES and IGFALS) associated with baseline cognitive stages and six proteins (ACES, C7, ZCD1, IL-17C, CC055, and SO5A1) linked to incident AD development. These longitudinal protein changes demonstrated significant associations with both neuroimaging and plasma AD biomarkers, achieving a 76% AUC-ROC for predicting incident AD.

Key Finding: The dynamic changes in plasma proteins over time provide valuable molecular signatures for AD progression and demonstrate strong potential as non-invasive biomarkers for predicting the development of Alzheimer's disease, offering a promising avenue for early detection and intervention strategies. 

Dr. Eun Hye Lee, a postdoctoral researcher in our lab, gave an oral presentation titled "DuAL-Net: A Hybrid Framework for AD Prediction from WGS via Local SNP Windows and Global Annotations" at the International Conference on Intelligent Biology and Medicine (ICIBM) 2025. The conference was held in Columbus, Ohio from August 3-5, 2025, with Dr. Lee's presentation taking place on August 5. The presentation introduced a deep learning framework that combines local SNP windows with global genetic annotations for Alzheimer's disease prediction using whole genome sequencing data. Dr. Lee completed the presentation successfully with considerable audience interest.

Our lab presented three posters at the Alzheimer's Association International Conference (AAIC) 2025 in Toronto.

 

1. Deep-Block: Large-scale WGS Analysis for Alzheimer's Disease Risk Variant Detection Using Deep Learning
   • Taeho Jo, Eun Hye Lee, Paula J Bice, Kwangsik Nho, Andrew J Saykin
      
2. Monte Carlo Dropout for Uncertainty-Aware Alzheimer's Disease Classification Using Transformer Models on Whole-Genome Sequencing Data
   • Taeho Jo, Eun Hye Lee
      
3. A Novel Deep Learning Model with Transformer Architectures to Enable Multi-scale Whole Genome Sequence Analysis for Alzheimer's Disease Dementia Prediction
   • Eun Hye Lee, Taeho Jo
      
   • 

   •  

The final installment of the four-part column series "AI in Medicine: From Nobel Discoveries to Clinical Frontiers" in the IU School of Medicine Research Updates blog. This column ties the series together, covering FDA-approved Alzheimer's treatments (lecanemab and donanemab), the challenge of early detection, and three AI-driven detection strategies: brain imaging analysis with CNNs, genomic analysis with the DeepBlock transformer tool, and metabolomic analysis with the c-SWAT tool.

Read the full column

The third installment of the four-part column series "AI in Medicine: From Nobel Discoveries to Clinical Frontiers" in the IU School of Medicine Research Updates blog. This column explores transformative "aha" moments in AI development, including the discovery of backpropagation, the 2020 scaling laws paper, and AlphaFold's breakthrough. It discusses Model Context Protocol enabling AI agent collaboration and identifies three irreplaceable human capacities: crafting narratives, formulating meaningful questions, and demonstrating authentic human connection.

Read the full column

The second installment of the four-part column series "AI in Medicine: From Nobel Discoveries to Clinical Frontiers" in the IU School of Medicine Research Updates blog. This column explains protein folding and its link to diseases like Alzheimer's and Parkinson's. It traces the CASP competition history, from David Baker's 2014 success to AlphaFold's breakthrough using CNNs, and then AlphaFold2's decisive victory using the Evoformer module, a Transformer variant.

Read the full column

The first installment of the four-part column series "AI in Medicine: From Nobel Discoveries to Clinical Frontiers" in the IU School of Medicine Research Updates blog. This column traces the history from the XOR problem through multi-layer perceptrons to modern deep learning, highlighting John Hopfield's Hopfield Network and Geoffrey Hinton's Boltzmann Machine as foundational contributions that earned the 2024 Nobel Prize in Physics.

Read the full column

Dr. Taeho Jo was invited to present at the AI-Biopharmaceutical Research Fusion Symposium hosted by Soongsil University. The symposium was held in Seoul, South Korea on May 30, 2025. His presentation, titled "How to Apply Nobel-Winning Deep Learning Breakthroughs to Alzheimer's Research," demonstrated practical approaches for applying Nobel Prize-winning deep learning techniques to Alzheimer's disease research, bridging the gap between cutting-edge AI advances and clinical applications.

The Indiana University (IU) School of Medicine webzine recently featured an interview discussing my research on utilizing AI for the early diagnosis and treatment of Alzheimer's disease. It also includes information regarding the recent REC Scholar Award.

You can read the full article via the link below.

https://medicine.iu.edu/blogs/research-updates/can-ai-revolutionize-alzheimers-care-iu-researcher-rec-scholar-award-winner-aims-for-earlier-detection-and-effective-treatments 

Dr. Taeho Jo spoke with DailyMedi, a Korean medical and health newspaper, to share insights on the current state and future prospects of medical AI. Over the course of 13 questions, he explored the latest developments, challenges, and opportunities shaping this rapidly evolving field.

Article (Korean)

Full Interview (Korean)

Dr. Taeho Jo has been selected as the 2025 Indiana Alzheimer's Disease Research Center (IADRC) Research Education Component (REC) Scholar.

Meghana Vodnala joined the lab as a research intern. With a Bachelor's degree in Dental Surgery and current pursuit of a Master's in Health Informatics, she will focus on studying data analytics and artificial intelligence.

Dr. Jo has been appointed as Associate Editor of Genomics & Informatics, a scientific journal published by Springer Nature Group, effective January 9, 2025. In this role, he will contribute to the editorial oversight of research publications in genomics and informatics.

"Linkage Disequilibrium-Informed Deep Learning Framework for Alzheimer's Disease"

This study introduces Deep-Block, a novel multi-stage deep learning framework that incorporates biological knowledge for analyzing large-scale genomic data in Alzheimer's disease. Applied to the Alzheimer's Disease Sequencing Project (ADSP) dataset of 7,416 participants, the framework successfully identified both known AD-associated variants (including APOE) and novel genetic loci. The findings were validated through brain tissue-specific eQTL analysis and cross-referenced with established AD genetics databases.

Key Innovation: Deep-Block's three-stage approach combines linkage disequilibrium patterns, sparse attention mechanisms, and advanced machine learning algorithms to effectively process large-scale genomic data while preserving critical biological relationships.  [LINK]

Received the Best Poster Award for the research "Identifying Alzheimer's-Related Genetic Regions Using Deep Learning on Whole-Genome Data" at the Asia-Pacific Psychiatry and Aging Conference.

Dr. Jo delivered his second lecture on EBS TV, Korea's leading public educational broadcaster. The presentation focused on the significance of 2024's Nobel Prize and current applications of AI in research, making complex scientific developments accessible to the general public.

Dr. Jo presented a hybrid lecture at Hallym University School of Medicine and Hallym Medical Center in Chuncheon, Korea. His presentation, "The Impact of Deep Learning's 2024 Nobel Prize: Innovations in Alzheimer's Prediction Through Multi-omics Analysis," examined the significance of AI's Nobel Prize recognition and showcased new developments using deep learning in Alzheimer's research.

Dr. Jo delivered a keynote presentation at the IUSM Imaging Sciences Research Seminar Series, exploring the transformative impact of deep learning on Alzheimer's research and beyond. The talk was hosted by the Indiana University School of Medicine Department of Radiology and Imaging Sciences in collaboration with the Indiana Alzheimer's Disease Research Center and Center for Neuroimaging on November 19, 2024.

Dr. Jo presented his research on multi-modal biomarker & genomic analysis using SWAT transformers at IUSM Conference - Neurosurgery Morning Grand Rounds, Indianapolis, Indiana

Dr. Eunhye Lee Joins Our Lab as a Postdoctoral Researcher. Dr. Lee, who earned her BS in Pharmacy, MD, and PhD from the same university and completed her Neurology Fellowship at a medical center, has joined our lab. She will focus on research involving AI applications in genomics and biological pathway analysis.

Dr. Jo was invited as a featured AI instructor at FastCampus, Korea's largest and premier tech education platform with over 70,000 active learners in paid courses. He delivered an in-depth curriculum covering the fundamentals of deep learning through advanced architectures including NN, MLP, RNN, CNN, GAN, Autoencoders and Transformer (Attention mechanisms, BERT, and GPT). 

Dr. Jo presented a research poster titled "Applying Deep Learning to Uncover Genetic Insights in Alzheimer's Disease: Analysis of 7,416 Whole Genome Sequences" at The 13th International Conference of Clinical Laboratory Automation in 2024. The presentation received the Silver Best Award for its innovative analysis of whole genome sequences using deep learning approaches. CopyRetry

Dr. Jo delivered a 6-hour intensive lecture at the International Healthcare Symposium, hosted by Ajou University Medical Center and the Graduate School of AI Convergence Innovation, Korea

Dr. Jo delivered an invited lecture at Seoul National University's College of Veterinary Medicine. His presentation covered the trends and research implications of deep learning/machine learning applications in Alzheimer's disease research, with a special focus on hyperparameter optimization techniques.

Dr. Jo delivered a comprehensive lecture "From Perceptron to Chat-GPT" at the 2023 AIAI Healthcare Symposium, jointly hosted by Ajou University Medical Center and Graduate School of AI Convergence Innovation. The presentation traced the evolution of deep learning from early neural networks to modern language models, focusing on their applications in healthcare.

Dr. Jo and colleagues published a research paper titled "Circular-SWAT for Deep Learning Based Diagnostic Classification of Alzheimer's Disease: Application to Metabolome Data" in eBioMedicine (doi: 10.1016/j.ebiom.2023.104820) in October 2023. The study introduced a novel Circular-Sliding Window Association Test (c-SWAT) methodology to improve classification accuracy in predicting Alzheimer's Disease using serum-based metabolomics data.

Dr. Jo presented his research at the Alzheimer's Association International Conference (AAIC) 2023. His poster presentation, "Deep Learning-based SWAT-Tab Approach for Identifying Genetic Variants using Whole Genome Sequencing," was co-authored with Kwangsik Nho and Andrew J. Saykin.

Dr. Jo presented his research at the Alzheimer's Association International Conference (AAIC) 2023. His poster presentation, "Deep Learning-based Integration of Neuroimaging and Genetic Data for Classification of Alzheimer's Disease," explored novel approaches to combining multi-modal data for improved AD diagnosis.

Dr. Jo served as a translator and technical reviewer for the "Beginner's Guide to Cognitive Neuroscience" (ISBN 979-1165219154), published by Gilbut Publishing in 2022. This 320-page introductory text provides aspiring neuroscientists with a comprehensive overview of cognitive neuroscience principles and foundational concepts.

Dr. Jo presented his research "Novel circular-SWAT for Deep Learning Based Diagnostic Classification of Alzheimer's Disease: Application to Metabolome Data" at the AI in Aging and Age-related Diseases conference (AIHC 2022) on November 11, 2022.

Dr. Jo delivered a presentation titled "A Dual-Deep Learning AI Strategy for Genetic Variants" at the Indiana Alzheimer's Disease Research Center Fall Research Symposium in 2022. The talk showcased innovative research applying dual-deep learning approaches to analyze genetic variants.

Dr. Jo presented "A Dual-Deep Learning AI Strategy to Identify Tau-associated Genetic Variants in Alzheimer's Disease" at the IADRC (Indiana Alzheimer's Disease Research Center) Annual All Hands Scientific Retreat & Reception on September 16, 2022.

Dr. Jo presented his research at the Alzheimer's Association International Conference (AAIC) 2022 in San Diego. His poster presentation, "Novel circular-SWAT for Deep Learning Based Diagnostic Classification of Alzheimer's Disease: Application to Metabolome Data," introduced a novel circular sliding window approach for analyzing metabolomics data using deep learning for AD classification.