Browsing by Author "Wang, T"

Now showing 1 - 4 of 4
  • Thumbnail Image
    ItemOpen Access
    FLAME: A Fast Large-scale Almost Matching Exactly Approach to Causal Inference
    (2017-08-01) Wang, T; Morucci, M; Awan, MU; Liu, Y; Roy, S; Rudin, C; Volfovsky, A
    A classical problem in causal inference is that of matching, where treatment units need to be matched to control units. Some of the main challenges in developing matching methods arise from the tension among (i) inclusion of as many covariates as possible in defining the matched groups, (ii) having matched groups with enough treated and control units for a valid estimate of Average Treatment Effect (ATE) in each group, and (iii) computing the matched pairs efficiently for large datasets. In this paper we propose a fast and novel method for approximate and exact matching in causal analysis called FLAME (Fast Large-scale Almost Matching Exactly). We define an optimization objective for match quality, which gives preferences to matching on covariates that can be useful for predicting the outcome while encouraging as many matches as possible. FLAME aims to optimize our match quality measure, leveraging techniques that are natural for query processing in the area of database management. We provide two implementations of FLAME using SQL queries and bit-vector techniques.
  • Thumbnail Image
    ItemOpen Access
    On enrichment strategies for biomarker stratified clinical trials
    (Journal of Biopharmaceutical Statistics, 2017-09-07) Wang, X; Zhou, J; Wang, T; George, SL
    In the era of precision medicine, drugs are increasingly developed to target subgroups of patients with certain biomarkers. In large all-comer trials using a biomarker strati ed design (BSD), the cost of treating and following patients for clinical outcomes may be prohibitive. With a fixed number of randomized patients, the efficiency of testing certain treatments parameters, including the treatment effect among biomarker positive patients and the interaction between treatment and biomarker, can be improved by increasing the proportion of biomarker positives on study, especially when the prevalence rate of biomarker positives is low in the underlying patient population. When the cost of assessing the true biomarker is prohibitive, one can further improve the study efficiency by oversampling biomarker positives with a cheaper auxiliary variable or a surrogate biomarker that correlates with the true biomarker. To improve efficiency and reduce cost, we can adopt an enrichment strategy for both scenarios by concentrating on testing and treating patient subgroups that contain more information about specifi c treatment parameters of primary interest to the investigators. In the first scenario, an enriched biomarker strati ed design (EBSD) enriches the cohort of randomized patients by directly oversampling the relevant patients with the true biomarker, while in the second scenario, an auxiliary-variable-enriched biomarker strati ed design (AEBSD) enriches the randomized cohort based on an inexpensive auxiliary variable, thereby avoiding testing the true biomarker on all screened patients and reducing treatment waiting time. For both designs, we discuss how to choose the optimal enrichment proportion when testing a single hypothesis or two hypotheses simultaneously. At a requisite power, we compare the two new designs with the BSD design in term of the number of randomized patients and the cost of trial under scenarios mimicking real biomarker strati ed trials. The new designs are illustrated with hypothetical examples for designing biomarker-driven cancer trials.
  • Thumbnail Image
    ItemOpen Access
    Structural Tolerance Factor Approach to Defect-Resistant I2-II-IV-X4 Semiconductor Design
    (Chemistry of Materials, 2020-02-25) Sun, JP; McKeown Wessler, GC; Wang, T; Zhu, T; Blum, V; Mitzi, DB
    Copyright © 2020 American Chemical Society. Recent work on quaternary semiconductors Cu2BaSn(S,Se)4 and Ag2BaSnSe4 for photovoltaic and thermoelectric applications, respectively, has shown the promise of exploring the broader family of defect-resistant I2-II-IV-X4 materials (where I, II, and IV refer to the formal oxidation state of the metal cations and X is a chalcogen anion) with tetrahedrally coordinated I/IV cations and larger II cations (i.e., Sr, Ba, Pb, and Eu) for optoelectronic and energy-related applications. Chemical dissimilarity among the II and I/IV atoms represents an important design motivation because it presents a barrier to antisite formation, which otherwise may act as electronically harmful defects. We herein show how all 31 experimentally reported I2-II-IV-X4 examples (with large II cations and tetrahedrally coordinated smaller I/IV cations), which form within five crystal structure types, are structurally linked. Based on these structural similarities, we derive a set of tolerance factors that serve as descriptors for phase stability within this family. Despite common usage in the well-studied perovskite system, Shannon ionic radii are found to be insufficient for predicting metal-chalcogen bond lengths, pointing to the need for experimentally derived correction factors as part of an empirically driven learning approach to structure prediction. We use the tolerance factors as a predictive tool and demonstrate that four new I2-II-IV-X4 compounds, Ag2BaSiS4, Ag2PbSiS4, Cu2PbGeS4, and Cu2SrSiS4, can be synthesized in correctly predicted phases. One of these compounds, Ag2PbSiS4, shows potentially promising optoelectronic properties for photovoltaic applications.
  • Thumbnail Image
    ItemOpen Access
    The macrophage: Switches from a passenger to a driver during anticancer therapy
    (Oncoimmunology, 2015) Wang, T; Feldman, GM; Herlyn, M; Kaufman, RE
    We have recently discovered that BRAF inhibitors induce potent macrophage responses that confer melanoma resistance to therapy. Our studies lay a foundation for the hypothesis that macrophages switch their role from a passenger to a driver for tumor survival during therapeutic treatment, suggesting that agents that target macrophages can be an important component of "cocktail" anticancer therapy.