About

My research involves the development and application of mathematical, computational, and statistical techniques to theoretical and methodological problems within the areas of social and biophysical network analysis, mathematical sociology, quantitative methodology, molecular modeling/analysis, and human judgment and decision making.

Research topics

• Medicine
• Biology
• Sociology
• Geography
• Mathematics
• Computer Science
• Data Mining
• Psychology
• Econometrics
• Environmental health
• Virology
• Machine Learning
• Chemistry
• Computational biology
• Social psychology
• Nursing
• Applied psychology
• Developmental psychology
• Demography
• Ecology
• Psychiatry
• Biochemistry
• Economics
• Genetics

Selected publications

• The Decay of Impact with Network Distance in Linear Diffusion Processes

  arXiv (Cornell University) · 2026-04-24

  preprintOpen accessSenior author

  Many processes related to status, power, and influence within social networks have been modeled using forced linear diffusion models; examples include the highly successful Friedkin-Johnsen model of social influence, the status/power scores of Katz and Bonacich, and the widely used network autocorrelation model. While a basic assumption of such models is that the impact of one individual on another through any given path falls exponentially with path length, the total impact of the first individual on the second involves contributions from walks of all lengths; thus, while total impact is expected to decline with network distance, the relationship is not trivial. Here, we provide an approximate solution for the total impact of one node on another as a function of network distance, showing that the total impact is given to first order by a product of eigenvector centrality scores together with an expression in terms of the graph spectrum (eigenvalues of the adjacency matrix) that falls exponentially with distance. We also show how this solution can be refined using higher-order eigenvectors of the adjacency matrix. A numerical study on interpersonal networks drawn from educational settings verifies an average exponential decline in impact strength under the linear diffusion model, and shows that the first-order eigenvector approximation can often be a good proxy for total impact as obtained from the exact solution. This suggests a simple model that can be used to approximate total impact for social influence or status processes in a range of settings.

  PublisherDOI

• Transition State Theory for Network Dynamics

  Open MIND · 2026-03-07

  preprint1st authorCorresponding

  Many classic questions of structural theory concern discrete changes, such as the formation or dissolution of groups, role turnover, or faction realignment. Here, we consider a basic framework combining prior work on change paths and recent advances in dynamic network modeling with ideas from transition state theory. This framework facilitates both characterizing the process of structural change and, in some cases, predicting it. Notably, this approach allows approximate prediction of network change from cross-sectional models, under limited assumptions regarding the underlying microdynamics. We apply this framework to a simple model of faction realignment in small groups, showing that the process through which realignment occurs can be well-predicted ex ante for a number of different network micro-processes.

  DOI

• Mimicking oxidative damage in γS-crystallin with site-specific incorporation of 5-hydroxytryptophan

  Biophysical Reports · 2026-01-18

  articleOpen access

  The human eye lens plays an essential role in vision by focusing light onto the retina. This transparent tissue consists of densely packed crystallin proteins that exhibit remarkable solubility despite minimal protein turnover. Post-translational modifications that accumulate over a lifetime can reduce crystallin solubility, resulting in the precipitation or phase separation of protein aggregates. Oxidation is a common type of modification that can cause such opacification of the lens, particularly in age-related cataract. Here, we study the oxidation of W163 in γS-crystallin, a structural lens protein that is particularly vulnerable to oxidative stress. We were motivated by previous findings reporting the oxidation of this residue in diseased and UV- and γ-irradiated samples. Using genetic code expansion (GCE), we incorporated an oxidation mimic, 5-hydroxytryptophan (5HTP), at position 163 of γS-crystallin (γS-W163(5HTP)). This subtle change in the structural and electronic properties of its side chain is hypothesized to destabilize the hydrophobic core of the C-terminal domain. γS-W163(5HTP) was characterized and compared to the wild-type (γS-WT). Although the overall fold and stability of the two proteins were comparable, the aggregation of γS-W163(5HTP) was triggered at notably lower temperatures compared to γS-WT. Subsequent investigation of this observation using both simulations and experiments suggests a potential mechanism for polymerization as well as oxidation-induced conformational changes that may cause susceptibility to thermal aggregation. Our findings highlight the utility of GCE platforms for systematically evaluating the impact of post-translational modifications on disease-related proteins.

  PublisherDOI

• Peer influence decay and behavioral diffusion in adolescent networks: A simulation approach

  Science · 2026-04-30

  article

  How far does peer influence spread through social networks before dissipating? This study investigates the diffusion of smoking behavior in adolescent friendship networks using longitudinal data from two schools ( n = 3154 students) in the National Longitudinal Study of Adolescent to Adult Health. Using Stochastic Actor–Oriented Models, we simulate interventions targeting heavy smokers using various strategies (random, in-degree, eigenvector centrality) and coverage (10 to 100%). A new exponential decay model quantifies influence attenuation, revealing indirect peer influences, or spillover effects, up to three steps from targets. Targeting 10 to 30% of central individuals maximizes smoking reductions, but gains plateau beyond 40 to 50% owing to network saturation. In our analyses, the denser network exhibits broader diffusion and slower decay than the larger, sparser network. This decay metric optimizes intervention design across diverse network structures.

  PublisherDOI

• Jasmonate signaling and prey nutrient availability trigger distinct biochemical responses in the <i>Drosera capensis</i> feeding cycle

  PLANT PHYSIOLOGY · 2026-01-17

  articleOpen access

  The Cape sundew (Drosera capensis) is a carnivorous plant native to South Africa. Central to its prey capture and digestive processes is a complex array of biochemical processes that trigger the production of enzymes and small molecules. These processes are in part activated by the release of jasmonic acid, a plant defense hormone repurposed as a prey detection signal. Here, we use RNASeq and untargeted metabolomics to study the response of D. capensis to feeding stimuli. We confirm the expression of genes encoding digestive proteins predicted in prior genomic work and show up- and down-regulation for a number of enzyme classes in response to jasmonic acid. Metabolomics experiments indicate that many small molecules produced during feeding depend on specific nutrient inputs from prey (and not merely a jasmonic acid stimulus). These results shed light on the molecular basis of plant carnivory and the recruitment of existing biochemical pathways to perform specialized functions in Caryophyllales carnivorous plants.

  PublisherOA PDFDOI

• A Behavioral Micro-foundation for Cross-sectional Network Models

  arXiv (Cornell University) · 2026-05-04

  preprintOpen access1st authorCorresponding

  Models for cross-sectional network data have become increasingly well-developed in recent decades, and are widely used. This has led to a growing interest in the connection between such cross-sectional models and the behavioral processes from which the corresponding networks were presumably generated. Here, we build on prior work in this area to present a behavioral micro-foundation for cross-sectional network models, based on a continuous time stochastic choice mechanism, that can accommodate highly general classes of cases (including agents who are not themselves in the network, and multilateral edge control). As we show, the equilibrium behavior of this process under appropriate conditions can be expressed in exponential family form, allowing estimation of individual preferences using existing methods; the graph potential separates naturally into a preference-based term reflecting agent utilities, and an entropic term reflecting the rules of tie formation. We illustrate our approach via an analysis of friendship in a professional organization, and modeling of phase transitions in the structure of small groups.

  PublisherDOI

• BPS2026 – Site-specific incorporation of 5-hydroxytryptophan mimics oxidative damage in a human eye lens protein

  Biophysical Journal · 2026-02-01

  article
  PublisherDOI

• The Decay of Impact with Network Distance in Linear Diffusion Processes

  ArXiv.org · 2026-04-24

  articleOpen accessSenior author

  Many processes related to status, power, and influence within social networks have been modeled using forced linear diffusion models; examples include the highly successful Friedkin-Johnsen model of social influence, the status/power scores of Katz and Bonacich, and the widely used network autocorrelation model. While a basic assumption of such models is that the impact of one individual on another through any given path falls exponentially with path length, the total impact of the first individual on the second involves contributions from walks of all lengths; thus, while total impact is expected to decline with network distance, the relationship is not trivial. Here, we provide an approximate solution for the total impact of one node on another as a function of network distance, showing that the total impact is given to first order by a product of eigenvector centrality scores together with an expression in terms of the graph spectrum (eigenvalues of the adjacency matrix) that falls exponentially with distance. We also show how this solution can be refined using higher-order eigenvectors of the adjacency matrix. A numerical study on interpersonal networks drawn from educational settings verifies an average exponential decline in impact strength under the linear diffusion model, and shows that the first-order eigenvector approximation can often be a good proxy for total impact as obtained from the exact solution. This suggests a simple model that can be used to approximate total impact for social influence or status processes in a range of settings.

  PublisherOA PDF

• Retrospective Network Imputation from Life History Data: The Impact of Designs

  UNC Libraries · 2026-04-15

  articleOpen accessSenior author

  Retrospective life history designs are among the few practical approaches for collecting longitudinal network information from large populations, particularly in the context of relationships like sexual partnerships that cannot be measured via digital traces or documentary evidence. While all such designs afford the ability to &ldquo;peer into the past&rdquo; vis-&agrave;-vis the point of data collection, little is known about the impact of the specific design parameters on the time horizon over which such information is useful. In this article, we investigate the effect of two different survey designs on retrospective network imputation: (1) intervalN, where subjects are asked to provide information on all partners within the past [Formula: see text] time units; and (2) lastK, where subjects are asked to provide information about their [Formula: see text] most recent partners. We simulate a &ldquo;ground truth&rdquo; sexual partnership network using a published model of Krivitsky (2012), and we then sample this data using the two retrospective designs under various choices of [Formula: see text] and [Formula: see text]. We examine the accumulation of missingness as a function of time prior to interview, and we investigate the impact of this missingness on model-based imputation of the state of the network at prior time points via conditional ERGM prediction. We quantitatively show that&mdash;even setting aside problems of alter identification and informant accuracy&mdash;choice of survey design and parameters used can drastically change the amount of missingness in the dataset. These differences in missingness have a large impact on the quality of retrospective parameter estimation and network imputation, including important effects on properties related to disease transmission.

  PublisherDOI

• Transition State Theory for Network Dynamics

  ArXiv.org · 2026-03-07

  articleOpen access1st authorCorresponding

  Many classic questions of structural theory concern discrete changes, such as the formation or dissolution of groups, role turnover, or faction realignment. Here, we consider a basic framework combining prior work on change paths and recent advances in dynamic network modeling with ideas from transition state theory. This framework facilitates both characterizing the process of structural change and, in some cases, predicting it. Notably, this approach allows approximate prediction of network change from cross-sectional models, under limited assumptions regarding the underlying microdynamics. We apply this framework to a simple model of faction realignment in small groups, showing that the process through which realignment occurs can be well-predicted ex ante for a number of different network micro-processes.

  PublisherOA PDF

Recent grants

• RAPID/Collaborative Research: Agency COVID-19 Risk Communication on Social Media: Characterizing Drivers of Message Retransmission and Engagement

  NSF · $98k · 2020–2022

• Collaborative Research: Informal Online Communication in Extreme Events: Content, Dynamics, and Structure

  NSF · $310k · 2010–2014

• Statistical Models for Dynamic Networks with Endogenous Vertex Migration

  NSF · $350k · 2018–2022

• DHB: Large-scale Spatially Embedded Interpersonal Networks: Measurement, Modeling, and Dynamics

  NSF · $749k · 2008–2014

• Doctoral Dissertation Research: Dynamic Network Models for the Scalable Analysis of Networks with Missing or Sampled Joint Edge/Vertex Evolution

  NSF · $15k · 2013–2014

Frequent coauthors

• Rachel W. Martin

  University of Oxford

  39 shared

• Zack W. Almquist

  University of Washington

  33 shared

• John R. Hipp

  University of California, Irvine

  32 shared

• Athina Markopoulou

  31 shared

• Jeannette Sutton

  University at Albany, State University of New York

  24 shared

• Pavel N. Krivitsky

  UNSW Sydney

  24 shared

• Minas Gjoka

  Google (United States)

  22 shared

• Maciej Kurant

  University of California, Irvine

  20 shared

Education

• Ph.D., Social and Decision Sciences

  Carnegie Mellon University

• M.S.

  Carnegie Mellon University

• B.S.

  Duke University

Awards & honors

• Leo A. Goodman Award
• Linton C. Freeman Award
• William Richards Awards
• Kavli Fellow
• Fellow of the AAAS