Shawn Mankad

Computational and Empirical Studies Predict Mycobacterium tuberculosis-Specific T Cells as a Biomarker for Infection Outcome

Identifying biomarkers for tuberculosis (TB) is an ongoing challenge in developing immunological correlates of infection outcome and protection. Biomarker discovery is also necessary for aiding design and testing of new treatments and vaccines. To effectively predict biomarkers for infection progression in any disease, including TB, large amounts of experimental data are required to reach statistical power and make accurate predictions. We took a two-pronged approach using both experimental and computational modeling to address this problem. We first collected 200 blood samples over a 2- year period from 28 non-human primates (NHP) infected with a low dose of Mycobacterium tuberculosis. We identified T cells and the cytokines that they were producing (single and multiple) from each sample along with monkey status and infection progression data. Machine learning techniques were used to interrogate the experimental NHP datasets without identifying any potential TB biomarker. In parallel, we used our extensive novel NHP datasets to build and calibrate a multi-organ computational model that combines what is occurring at the site of infection (e.g., lung) at a single granuloma scale with blood level readouts that can be tracked in monkeys and humans. We then generated a large in silico repository of in silico granulomas coupled to lymph node and blood dynamics and developed an in silico tool to scale granuloma level results to a full host scale to identify what best predicts Mycobacterium tuberculosis (Mtb) infection outcomes. The analysis of in silico blood measures identifies Mtb-specific frequencies of effector T cell phenotypes at various time points post infection as promising indicators of infection outcome. We emphasize that pairing wetlab and computational approaches holds great promise to accelerate TB biomarker discovery.

Biclustering Three-Dimensional Data Arrays With Plaid Models

Three-dimensional data arrays (collections of individual data matrices) are increasingly prevalent in modern data and pose unique challenges to pattern extraction and visualization. This article introduces a biclustering technique for exploration and pattern detection in such complex structured data. The proposed framework couples the popular plaid model together with tools from functional data analysis to guide the estimation of bicluster responses over the array. We present an efficient algorithm that first detects biclusters that exhibit strong deviations for some data matrices, and then estimates their responses over the entire data array. Altogether, the framework is useful to home in on and display underlying structure and its evolution over conditions/time. The methods are scalable to large datasets, and can accommodate a variety of dynamic patterns. The proposed techniques are illustrated on gene expression data and bilateral trade networks. Supplementary materials are available online.

Discovery of path-important nodes using structured semi-nonnegative matrix factorization

Identifying critical components in networked systems is a key problem for many important applications in a diverse set of fields, including epidemiology, e-commerce and traffic systems. This paper describes the development and application of a semi-nonnegative matrix factorization for structural discovery featuring nodes that are important for transmission over social networks. The technique allows the practitioner to perform structured matrix factorization by specifying context-specific network statistics that guide the solution. The techniques are demonstrated on a network derived from Twitter data.

Measuring Influence of Users in Twitter Ecosystems Using a Counting Process Modeling Framework

Data extracted from social media platforms are both large in scale and complex in nature, since they contain both unstructured text, as well as structured data, such as time stamps and interactions between users. A key question for such platforms is to determine influential users, in the sense that they generate interactions between members of the platform. Common measures used both in the academic literature and by companies that provide analytics services are variants of the popular web-search PageRank algorithm applied to networks that capture connections between users. In this work, we develop a modeling framework using multivariate interacting counting processes to capture the detailed actions that users undertake on such platforms, namely posting original content, reposting and/or mentioning other users’ postings. Based on the proposed model, we also derive a novel influence measure. We discuss estimation of the model parameters through maximum likelihood and establish their asymptotic properties. The proposed model and the accompanying influence measure are illustrated on a dataset covering a five-year period of the Twitter actions of the members of the U.S. Senate, as well as mainstream news organizations and media personalities. Supplementary material is available online including computer code, data, and derivation details.

Analysis of Multiview Legislative Networks with Structured Matrix Factorization: Does Twitter Influence Translate to the Real World?

The rise of social media platforms has fundamentally altered the public discourse by providing easy to use and ubiquitous forums for the exchange of ideas and opinions. Elected officials often use such platforms for communication with the broader public to disseminate information and engage with their constituencies and other public officials. In this work, we investigate whether Twitter conversations between legislators reveal their real-world position and influence by analyzing multiple Twitter networks that feature different types of link relations between the Members of Parliament (MPs) in the United Kingdom and an identical data set for politicians within Ireland. We develop and apply a matrix factorization technique that allows the analyst to emphasize nodes with contextual local network structures by specifying network statistics that guide the factorization solution. Leveraging only link relation data, we find that important politicians in Twitter networks are associated with real-world leadership positions, and that rankings from the proposed method are correlated with the number of future media headlines.

Visual Analytics for Network-Based Market Surveillance

Time series of graphs are increasingly prevalent in modern economic and financial data and pose unique challenges to visual exploration and pattern extraction. This paper describes the application of matrix factorizations that enhance existing visualization techniques for exploration and pattern detection in graph time-series. The combination of matrix factorization and visualizations allows the user to home in on and display interesting, underlying structure and its evolution over time. The methods are scalable to data sets with a large number of time points or nodes, and can accommodate sudden changes to graph topology. The tools are used to summarize how dynamics in the interbank and equity markets changed during the sub-prime crisis for banks in the Eurozone area.

Balance Sheet Driven Probability Factorization for Inferring Bank Holdings: Extended Abstract

 Assistant Professor of Operations, Technology and Information Management, Samuel Curtis Johnson Graduate School of Management, Cornell University, 2015 – Present o Graduate Field Member in Statistics, 2017 – Present  Assistant Professor of Business Analytics, Robert H. Smith School of Business, University of Maryland, 2013 – 2015 o Affiliate Faculty of Applied Mathematics and Scientific Computation, University of Maryland, 2014 – 2015  Federal Contractor, the U.S. Commodity Futures Trading Commission, 2009 – 2013  Dissertation Intern, Federal Reserve Board of Governors, Summer 2012

A Feedback and Evaluation System That Provokes Minimal Retaliation by Trainees

Background: Grade inflation is pervasive in educational settings in the United States. One driver of grade inflation may be faculty concern that assigning lower clinical performance scores to trainees will cause them to retaliate and assign lower teaching scores to the faculty member. The finding of near-zero retaliation would be important to faculty members who evaluate trainees. Methods: The authors used a bidirectional confidential evaluation and feedback system to test the hypothesis that faculty members who assign lower clinical performance scores to residents subsequently receive lower clinical teaching scores. From September 1, 2008, to February 15, 2013, 177 faculty members evaluated 188 anesthesia residents (n = 27,561 evaluations), and 188 anesthesia residents evaluated 204 faculty members (n = 25,058 evaluations). The authors analyzed the relationship between clinical performance scores assigned by faculty members and the clinical teaching scores received using linear regression. The authors used complete dyads between faculty members and resident pairs to conduct a mixed effects model analysis. All analyses were repeated for three different epochs, each with different administrative attributes that might influence retaliation. Results: There was no relationship between mean clinical performance scores assigned by faculty members and mean clinical teaching scores received in any epoch (P ≥ 0.45). Using only complete dyads, the authors’ mixed effects model analysis demonstrated a very small retaliation effect in each epoch (effect sizes of 0.10, 0.06, and 0.12; P ⩽ 0.01). Conclusions: These results imply that faculty members can provide confidential evaluations and written feedback to trainees with near-zero impact on their mean teaching scores.

Two-Stage Plans for Estimating the Inverse of a Monotone Function

This study investigates two-stage plans based on nonparametric procedures for estimating an inverse regression function at a given point. Specifically, isotonic regression is used at stage one to obtain an initial estimate followed by another round of isotonic regression in the vicinity of this estimate at stage two. It is shown that such two-stage plans accelerate the convergence rate of one-stage procedures and are superior to existing two-stage procedures that use local parametric approximations at stage two when the available budget is moderate and/or the regression function is “ill-behaved.” Both Wald- and likelihood ratio-type confidence intervals for the threshold value of interest are investigated and the latter are recommended in applications due to their simplicity and robustness. The developed plans are illustrated through a comprehensive simulation study and an application to car fuel efficiency data.

Do U.S. Regulators Listen to the Public? Testing the Regulatory Process with the RegRank Algorithm

We propose a tool called RegRank that can be used to measure and test whether government regulatory agencies adjust aspects of final rules in response to comments received from the public. The algorithm, which combines customized dictionaries with LDA topic models, is used to analyze the text of public rulemaking documents of the Commodity Futures Trading Commission (CFTC) - a federal regulatory agency in charge of implementing parts of the Dodd-Frank Wall Street Reform and Consumer Protection Act. A key finding based on the available data is that the government adjusts its final rules in the direction of public comments.