Building a Personal Research Pipeline with AI
An automated tool for fetching, scoring, and summarizing daily academic literature across Nature, Science, bioRxiv, and arXiv.
Biography
Dr. Wei Zhang is a Postdoctoral Associate in the Division of Biostatistics, Department of Public Health Sciences at the University of Miami Miller School of Medicine. Her research develops statistical machine learning methods for integrative multi-omics analysis and biomarker discovery across neurodegenerative disease and cancer.
Research
Current research spans four methodological and translational threads:
- Multivariate random forest methods — for matched multi-omics integration, clustering, and feature selection.
- Epigenetic biomarkers of aging and Alzheimer's disease — blood and brain DNA methylation signatures of cognitive decline, resilience, and dementia risk.
- Integrative epigenetic–neurostructural age clocks — combining blood DNA methylation with brain MRI for Alzheimer's disease prognosis.
- Predictive models for treatment response in oncology — matched tumor–organoid transcriptomes and network-based biomarker selection in colorectal and breast cancer.
Active Projects
Multivariate Random Forest Framework for Multi-omics Integration
A directed random forest framework that learns sample-by-sample similarity from shared terminal-node structure across omics blocks, then decomposes the result into shared and omics-specific components. Two sub-projects extend the framework in complementary directions:
- a. Variable Selection Robust biomarker ranking via an Inverse-Minimal-Depth (IMD) importance measure, with three selection strategies benchmarked on pan-cancer TCGA data and applied to dementia prediction in ADNI. GitHub → Project page →
- b. Clustering A forest-based decomposition that separates shared cross-modal structure from modality-specific residual signal — letting each be clustered on its own terms. Applied to TCGA head-and-neck cancer and to matched blood DNAm + structural MRI in ADNI. Preprint → GitHub → R package → Vignette → Project page →
Integrative Aging Biomarker of Blood DNA Methylation and Brain MRI for AD Prognosis
A disease-focused integrative epigenetic and neurostructural age clock for Alzheimer's risk stratification. PI: Wei Zhang; Mentor: Lily Wang, Ph.D.; Co-Mentors: Tatjana Rundek, M.D., Ph.D., and David Loewenstein, Ph.D. Proposed under NIH K99/R00 and AARFA.
Epigenetic Biomarkers for Alzheimer's and Cognitive Health
Discovery and validation of blood-based DNA methylation signatures of incident dementia, lifestyle-based cognitive resilience, and the aging epigenome in longitudinal cohorts.
Prediction Models for Cancer Biomarkers
Transcriptome-based models for chemotherapy response using matched colorectal tumor–organoid expression profiles, with network-based biomarker selection and meta-analyses in triple-negative breast cancer.
Selected Publications
Selected peer-reviewed articles and preprints. Click any entry to expand. Full list on Google Scholar.
-
1.Zhang W, Wang L, Franzmann EJ, Chen XS. (2026). Multivariate Random Forests for Cross-Modal Multi-Omics Integration. bioRxiv, preprint.+Abstract. multiRF is a random-forest-based method that separates shared cross-modal structure from modality-specific residual signal in multi-omics data. It recovers shared clusters as well or better than established integrative methods under nonlinear structure; applied to TCGA head-and-neck cancer and to matched ADNI blood DNAm + structural MRI it recovers a common disease axis while preserving biology unique to a single data type.
-
2.Zhang W, Lukacsovich D, Young JI, Gomez L, Schmidt MA, Kunkle B, Chen XS, Martin ER, Wang L. (2025). The Aging Epigenome: Integrative Analyses Reveal Functional Overlap with Alzheimer's Disease. GeroScience, accepted.+Abstract. Integrative analyses of aging and AD methylomes uncover shared loci and functional pathways, suggesting overlapping epigenetic programs between normal aging and neurodegeneration.
-
3.Zhang W, Huang H, Wang L, Lehmann BD, Chen XS. (2025). An Integrative Multi-Omics Random Forest Framework for Robust Biomarker Discovery. GigaScience, giaf148.+Abstract. We introduce a multivariate random forest framework that jointly models multiple omics modalities, yielding robust biomarker rankings and improved prediction compared to single-omic baselines across benchmark cohorts.
-
4.Zhang W, Lukacsovich D, Young JI, Gomez L, Schmidt MA, Martin ER, Kunkle BW, Chen X, O'Shea DM, Galvin JE, Wang L. (2025). DNA Methylation Signature of a Lifestyle-based Resilience Index for Cognitive Health. Alzheimer's Research & Therapy, 17, 88.+Abstract. A blood DNA-methylation signature captures lifestyle-based cognitive resilience and predicts trajectories of cognitive decline independent of known risk factors.
-
5.Zhang W, Young JI, Gomez L, Schmidt MA, Lukacsovich D, Kunkle B, Chen XS, Martin ER, Wang L. (2025). Blood DNA Methylation Signature for Incident Dementia: Evidence from Longitudinal Cohorts. Alzheimer's & Dementia, 21:e14496.+Abstract. Using multiple longitudinal cohorts we derive and validate a blood-based methylation signature that stratifies risk of incident dementia years before diagnosis.
-
6.Zhang W, Wu C, Huang H, Bleu P, Zambare W, Alvarez J, Wang L, Paty PB, Romesser PB, Smith JJ, Chen XS. (2025). Enhancing Chemotherapy Response Prediction via Matched Colorectal Tumor-Organoid Gene Expression Analysis and Network-Based Biomarker Selection. Translational Oncology, 52:102238.+Abstract. Matched tumor–organoid transcriptomes improve chemotherapy response prediction in colorectal cancer, supporting organoid-guided precision treatment.
-
7.Zhang W, Young JI, Gomez L, Schmidt MA, Lukacsovich D, Varma A, Chen XS, Kunkle B, Martin ER, Wang L. (2024). Critical Evaluation of the Reliability of DNA Methylation Probes on the Illumina MethylationEPIC v1.0 BeadChip Microarrays. Epigenetics, 19(1):2333660.+Abstract. A systematic evaluation of probe reliability on the EPIC v1.0 BeadChip identifies probe subsets with variable reproducibility and offers filtering recommendations.
-
8.Zhang W, Young JI, Gomez L, Schmidt MA, Lukacsovich D, Varma A, Chen XS, Martin ER, Wang L. (2023). Distinct CSF Biomarker-Associated DNA Methylation in Alzheimer's Disease and Cognitively Normal Subjects. Alzheimer's Research & Therapy, 15:78.+Abstract. CSF biomarker-associated methylation patterns differ between AD and cognitively normal subjects, highlighting disease-specific epigenetic regulation.
-
9.Zhang W, Li E, Wang L, Lehmann BD, Chen XS. (2023). Transcriptome Meta-Analysis of Triple-Negative Breast Cancer Response to Neoadjuvant Chemotherapy. Cancers, 15(8):2194.+Note. Transcriptome meta-analysis of triple-negative breast cancer response to neoadjuvant chemotherapy.
Software & Open-Source Tools
multiRF
An R package for integrating matched multi-omics datasets with multivariate random forests. Fits directed forest models across omics blocks, learns sample-by-sample similarity from shared terminal-node structure, and decomposes the result into shared and omics-specific components for clustering, variable selection, and visualization.
Presentations
-
A Disease-Focused Integrative Epigenetic and Neurostructural Age Clock for Alzheimer's Risk
-
Multivariate Random Forest-Based Clustering for Integrative Multi-Omics Analysis
-
Multivariate Random Forest-Based Clustering for Integrative Multi-Omics Analysis
-
An Integrative Multi-Omics Random Forest Framework for Robust Biomarker Discovery
-
An X Chromosome-Wide DNA Methylation Study of Alzheimer's Disease
-
Unlocking the Potential of Multi-Omics Data Integration Using Multivariate Random Forest Approach
-
Distinct CSF Biomarker-Associated DNA Methylation in Alzheimer's Disease and Cognitively Normal Subjects
-
Iterative Multivariate Random Forest for Feature Selection in Integrating Multi-Omics Datasets
Teaching
-
EPH 705: Advanced Statistical Methods — Teaching Assistant
-
STAT 6201: Applied Linear Models — Teaching Assistant
Honors & Awards
-
Award of Academic Merit
-
Best Student Poster Award
-
Travel Award
Education
-
Ph.D. in Biostatistics
-
M.S. in Statistics
-
B.S. in Economics Analysis & Actuarial Mathematics
Professional Development
-
Duke EHR Study Design Workshop
-
Code Rigor and Reproducibility with R Boot Camp
Professional Service
Peer review for:
Nature Communications · Neurobiology of Aging · Statistics in Medicine · Scientific Reports · Discover Applied Sciences · Biology Direct · Discover Oncology · Medicine in Omics.
Other service:
Reviewer, ASA South Florida Student Data Challenge (2026).
Memberships:
International Biometric Society (ENAR) · American Statistical Association (ASA) · International Society to Advance Alzheimer's Research and Treatment (ISTAART).
Blog & Side Projects
Technical notes on tools and side projects built alongside research.
Bloomly: An Atomic Knowledge Hub
A personal knowledge app to capture atomic notes, link them into a network, and use AI to help them grow. Available on web and iOS.
Correspondence
Open to research collaborations in computational biology, biostatistics, and statistical methodology. Please direct inquiries to wei.zhang60@med.miami.edu.
Updated April 2026