Tianhua Zhai

Dr. Tianhua Zhai

orcid

Research Focus & Metrics

Academic Career Length

3Years (2024-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
13 May 2024
Verified
Conference Paper
Open access
CHChi-Yang Hsu
Chi-Yang Hsu
KCKyle Cox
Kyle Cox
JXJiawei Xu
Jiawei Xu
ZTZhen Tan
Zhen Tan
TZTianhua Zhai
Tianhua Zhai
YDYing Ding
Ying Ding

4 December 2024 in Targets
Verified
Journal Article
Open access
PDF file
CSCatherine Sharo
Catherine Sharo
JZJiayu Zhang
Jiayu Zhang
TZTianhua Zhai
Tianhua Zhai
JBJingxuan Bao
Jingxuan Bao
AGAndrés Garcia-Epelboim
Andrés Garcia-Epelboim
Zuyi HuangZuyi Huang
Zuyi Huang
Alzheimer’s disease is a neurodegenerative disease that continues to have a rising number of cases. While extensive research has been conducted on Alzheimer’s disease in the last few decades, only a few drugs have been approved by the FDA for its treatment, and even fewer aim to be curative rather than manage symptoms. There remains an urgent need to understand disease pathogenesis, as well as identify new targets for further drug discovery. Alzheimer’s disease (AD) is known to stem from the build-up of amyloid beta (Aβ) plaques, as well as tangles of tau proteins. Furthermore, inflammation in the brain is known to arise from the degeneration of tissue and the build-up of insoluble material. Therefore, there is a potential link between the pathology of AD and inflammation in the brain, especially as the disease progresses to later stages, where neuronal death and degeneration levels are higher. Proteins that are relevant to both brain inflammation and AD, thus, make ideal potential targets for therapeutics; however, the proteins need to be evaluated to determine which targets would be ideal for potential drug therapeutic treatments, or ‘druggable’ targets. Druggability analysis was conducted using two structure-based methods (i.e., drug-like density analysis and SiteMap), as well as a sequence-based approach, SPIDER. The most druggable targets were then evaluated using single-nucleus sequencing data for their clinical relevance to inflammation in AD. For each of the top five targets, small molecule docking was used to evaluate which FDA approved drugs were able to bind with the chosen proteins. The top targets included DRD2 (inhibits adenylyl cyclase activity), C9 (binds with C5B8 to form the membrane attack complex), C4b (binds with C2a to form C3 convertase), C5AR1 (a GPCR that binds C5a), and GABA-A-R (the GPCR involved in inhibiting neurotransmission). Each target had multiple potential inhibitors from the FDA-approved drug list with decent binding infinities. Among these inhibitors, two drugs were found to be top inhibitors for more than one protein target. They were C15H14N2O2 and v316 (paracetamol), originally used to treat pain/inflammation for cataracts and relieve headaches/fever, respectively. These results provide the groundwork for further experimental investigations or clinical trials.

23 September 2025 in Innovation in Aging
Verified
Journal Article
Open access
TSTianqi Shang
Tianqi Shang
SYShu Yang
Shu Yang
TZTianhua Zhai
Tianhua Zhai
WHWeiqing He
Weiqing He
EMElizabeth Mamourian
Elizabeth Mamourian
Li ShenLi Shen
Li Shen
Background and Objectives: Alzheimer’s disease (AD) and AD-related dementias (ADRD) are expected to affect over 100 million people by 2050, placing a significant strain on public health systems. Social determinants of health (SDoH), which include factors such as socioeconomic conditions and environment, play a crucial role in AD risk. Despite growing evidence, the understanding of SDoH’s impact on AD remains limited. Research Design and Methods: This study leverages large language models and knowledge graphs (KGs) to extract AD-related SDoH knowledge from literature and electronic health records (EHR). We integrate this knowledge into biological research on AD through KG construction and graph deep learning, performing KG-link predictions validated by multimodal biological data from single-cell RNA-seq and proteomics. Results: We generated an SDoH knowledge graph with around 92k triplets, integrating literature and EHR data. In various link prediction experiments, we observed higher accuracy when integrating SDoH into knowledge graphs. Additionally, exploratory predictions uncovered potential SDoH-gene interactions, many of which were validated through differential expression analysis using proteomics and RNA-seq data. Discussion and Implications: This novel KG-based analysis enhances link prediction in AD-related biomedical networks by integrating SDoH and biological knowledge. Our findings highlight the potential interaction between social determinants and biological factors in AD, offering insights into more personalized and socially aware healthcare interventions.

25 December 2025 in Alzheimer's & Dementia
Verified
Journal Article
Open access
TZTianhua Zhai
Tianhua Zhai
DSDavid Shen
David Shen
MNMagally Arevalo Navarro
Magally Arevalo Navarro
ZHZuyi Jacky Huang
Zuyi Jacky Huang
LSLi Shen
Li Shen
Background: Beta‐secretase 1 (BACE1) plays a crucial role in the formation of neurotoxic amyloid‐beta peptides in Alzheimer's disease. Effective BACE1 inhibitors are urgently needed for AD intervention. This study utilized available BACE1 inhibitor data to develop a machine‐learning approach that integrates both BACE1‐ligand interaction features and inhibitor QSAR descriptors to discover new BACE1 inhibitors. Method: Ligand‐protein interactions of 284 BACE1‐ligand complexes, along with experimental pIC50 values (indicating binding affinities), were obtained from the Protein Data Bank. Docking simulations were performed using Molsoft ICM software to reproduce the binding conformation of each BACE1‐ligand complex, yielding binding interaction features (seven force‐field energy components, such as hydrogen bonding and hydrophobic interactions). 881 QSAR features for each ligand were extracted using the PubChem molecular fingerprint calculator. Subsequently, 12 highly correlated QSAR features were selected via statistical tests on training data. Finally, 43 machine‐learning models were developed to predict pIC50 using three input sets: 1) binding interaction features, 2) the top 12 QSAR features identified, and 3) the combined 19 binding and QSAR features. Each model was trained on 80% of the data and tested on 20%, with performance evaluated on 100 random partitions using the R² metric. Result: The top 5 machine‐learning models demonstrated improved prediction accuracy when combining binding and QSAR features compared to using either feature set alone (Figure 1). The NuSVR model achieved the best prediction performance. In the best‐performing trials, NuSVR achieved R² = 0.78 using combined features, compared to R² = 0.65 and R² = 0.64 using binding and QSAR features, respectively (Figure 2). Additionally, the fuzzy Gaussian regression model with seven binding interaction features and molecular weight also demonstrated acceptable performance, with R² = 0.642 and mean squared error (MSE) = 0.45 (Figure 3). Conclusion: Machine‐learning models integrating binding and QSAR features enhance the prediction of BACE1‐ligand binding affinity (pIC50), enabling efficient virtual screening for potent BACE1 inhibitors. These advancements offer potential for accelerating drug discovery for Alzheimer’s disease. Future work will expand these models to include broader chemical databases, furthering intervention strategies for AD.

30 June 2026
Verified
Conference Paper
Open access
LZLening Zhao
Lening Zhao
TZTianhua Zhai
Tianhua Zhai
Li ShenLi Shen
Li Shen
Identifying druggable pockets in proteins is central to structure-based drug discovery, yet conventional ligand-binding-site prediction and cryptic-pocket prediction are typically treated as separate tasks requiring distinct tools, preprocessing pipelines, and evaluation protocols. This separation is limiting because the two problems share a common biological foundation: co-evolutionary patterns encoded in protein sequences carry implicit signals about both conventional and cryptic binding sites. We present UniPocket, a unified multitask framework for residue-level prediction of both ligand-binding and cryptic-pocket residues within a single shared-backbone architecture. UniPocket uses frozen per-residue ESM-2 embeddings as input to a lightweight residual MLP with two task-specific heads, trained on ligand-contact labels from recently deposited PDB structures and cryptic-pocket annotations from CryptoBench. Training alternates between ligand-labeled and cryptic-labeled mini-batches, while an orthogonality regularizer encourages the two heads to learn complementary rather than redundant signals. UniPocket achieves a macro ROC-AUC of 0.82 for cryptic-pocket prediction and 0.76 for ligand-binding-site prediction — matching or exceeding all dedicated single-task baselines on both tasks simultaneously, without any 3D structural input at inference time.

14 July 2026 in bioRxiv
Verified
Preprint
Open access
PDF file
CZCharles Zheng
Charles Zheng
TZTianhua Zhai
Tianhua Zhai
FZFangyuan Zhang
Fangyuan Zhang
Li ShenLi Shen
Li Shen
Lipid-processing microglia are among the microglial states most strongly associated with Alzheimer’s disease (AD) pathology, yet whether this association reproduces across independent cohorts, what transcriptional programs define the state, and which upstream signals and small molecules can modulate it remain unsolved. We address these questions through a cross-cohort analysis of one such substate (MG4) by integrating differential expression, transcription factor activity inference, gene set enrichment, and cell-cell communication across five independent single-nucleus RNA sequencing cohorts ( n total = 140 donors), with paired single-nucleus ATAC sequencing in one multi-omic cohort for epigenomic corroboration. A held-out cohort ( n = 150 donors) supported donor-level regression of MG4 proportion on ligand expression, and two spatial transcriptomics datasets ( n total = 30 donors) related ligand expression to MG4 identity in neighboring spots. MG4 was reproducibly enriched in AD across all five cohorts (pooled log 2 fold change = 0.90 , p = 3.0 × 10 −4 ). Expression-based inference and motif accessibility jointly nominated MITF and BACH1 as regulators of a program led by V-ATPase-driven lysosomal acidification and cholesterol efflux, a lysosomal-biogenesis signature distinct from the catabolic DAM and lipid-storage LDAM programs, with AD-specific upregulation of energy metabolism. FGF1 and TGFB2 were the most supported candidate upstream ligands, each significant in donor-level regression with further spatial evidence. Computational drug repurposing nominated ten blood-brain barrier-penetrant compounds as perturbational probes. Together, these results advance a described disease-associated microglial state into a reproducible, mechanistically framed regulatory model, providing candidate regulators, upstream ligands, and pharmacological probes for functional validation.