Danni Wu Email & Phone Number

Boston, Massachusetts, United States

Research Associate in the Department of Biostatistics at the Harvard T. H. Chan School of Public Health.Project: Comparing the effect of Dolutegravir and Efavirenz treatment on maternal postpartum depression in IMPAACT 2010 Phase III Study• Completed the statistical analysis plan and finalized the statistical analysis report• 1 first-authored abstract was submitted to CROI 2025 and 1 manuscript is in progressProject: Fusion Design to Combine Randomized Controlled Trials with Real-World Evidence• Developed a simulation framework to implement the fusion design among pregnant people living with HIV using causal inference methods (e.g., Inverse Odds of Sampling Weights)• Conducted simulations to evaluate the operating characteristics of fusion design• Collaborating with interdisciplinary teams and 3 manuscripts are in preparationProject: Phase IV Study of the Pharmacokinetics and Safety of Ceftriaxone and Benzathine Penicillin G During Pregnancy• Served as statistician for study design, study schema and protocol development

New York, New York, United States

The following projects are funded by NIH UL1TR001445.Project: Prospective Meta-analysis for COVID-19 • Collaborated with interdisciplinary teams on a prospective meta-analysis of 8 international COVID-19 trials, influencing FDA’s policy on COVID-19 convalescent plasma (CCP) use • Published 5 articles in top-tier journals (1 first-authored paper); presented at 7 international conferences• Led efforts to publish a Bayesian adaptive design framework guiding prospective meta-analysis statistical design• Conducted simulations to assess trial operating characteristics using high-performance computing cluster• Created Bayesian hierarchical models and stopping rules for nearly continuous data monitoring to assess efficacy• Designed a data pipeline using R and Stan to automate data visualization and interim analysis report generationProject: Bayesian Multivariate Model for Precision Medicine • Developed a Bayesian multivariate model to improve individualized treatment recommendations (ITRs) using mixed types of health outcomes• Reduced erroneous ITRs by 10% versus standard univariate models demonstrated by 30K+ simulated trials• Implemented model in COVID-19 trials, enhancing the estimation of treatment effect and optimal ITRsProject: Dose Effect Analysis of Convalescent Plasma • Conducted subgroup analysis using Generalized Additive Modeling (GAM) to examine the association between health outcomes and the dose across different levels of baseline factors• Combined GAM regression results across multiple imputed data sets using the median p-value rule

Cambridge, Massachusetts, United States

• Designed a Bayesian adaptive seamless Phase 2/3 trial for individualized neoantigen therapy• Designed a framework for early decisions using ctDNA clearance as a surrogate endpoint for survival outcomes• Provided statistical consultation on trial design, including extensive simulations to assess operating characteristics• Quantified gains and risks of proposed designs in comparison to standard frequentist designs and made decisions

Greater New York City Area

Supervisor: Dr. Jessica Anker• Conducted interrupted time series analysis on 40K+ electronic health records to estimate the effect of an electronic health record prescribing redesign on both opioid prescribing choices and keystrokes• Presented project updates biweekly to three cross-disciplinary professors to inform decision-making

Greater New York City Area

Supervisor: Dr. Samprit Banerjee• Compared the performance of 6 machine learning methods for predicting cardiac arrhythmia in 4K+ patients• Developed and applied a penalized Cox proportional hazards model with Lasso penalty• Improved accuracy by 27% over the previously published result

Ithaca, New York Area

• Supervisor: Dr. Kilian Q. Weinberger• Developed a Python library utilizing Gaussian process spectral mixture kernel regression• Devised a hybrid approach by combining gradient descent and selective acceptance of local optima,leading to a 40% increase in prediction accuracy

Beijing, China

• Conducted a comprehensive literature review on Machine Learning algorithms • Led a machine learning workshop

Beijing City, China

Time series analysis

China

• Wrote a grant proposal outlining the economic development impact of high-tech employees • Successfully collected, analyzed, and interpreted data spanning from 2000 to 2015, resulting in a top return offer.

Doctor Of Philosophy - Phd, Biostatistics

Master'S Degree, Biostatistics

Bachelor Of Science - Bs, Statistics