Erina Paul
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Erina Paul has a total of 33 co-authors. Here are the top co-authors by number of shared publications:
- Himel Mallick
Cornell University
6 shared publications - Piyali Basak
Biostatistics and Research Decision Sciences, Merck & Co., Inc., Rahway, NJ, USA
5 shared publications - Chitrak Banerjee
Analytical AI, Bristol Myers Squibb, Princeton, NJ, USA
2 shared publications - Matthew Evan Sprong
Governors State University, University Park, IL, USA
2 shared publications - Saptarshi Roy
Department of Statistics, Texas A&M University, College Station, TX, USA
2 shared publications - Sreya Sarkar
Department of Statistics and Actuarial Science, The University of Iowa, Iowa City, IA, USA
2 shared publications
Latest Publications
- Open Access
Conformalized Multiview Learning
S ummary: There is an overwhelmingly large body of literature and numerous algorithms already available on “multimodal AI”, based on different modeling and fusion paradigms. As multimodal AI becomes fully integrated into healthcare applications, the ability to say “I’m not sure” or “I don’t know” when uncertain is a necessary capability for safe clinical deployment, requiring the augmentation of current multimodal AI techniques. Our goal in this work is not to add yet another method to the growing toolbox of multimodal AI, but rather to (i) clarify how existing supervised multimodal AI methods can be understood through the lens of uncertainty quantification, and (ii) develop a unified framework that provides valid and interpretable uncertainty estimates for multimodal AI practitioners. We introduce Coracle, a conformalized framework for multimodal AI that seamlessly adapts to early, late, and intermediate fusion. Coracle provides theoretical marginal confidence guarantees and achieves valid finite-sample coverage without relying on distributional assumptions. Through extensive simulations and analyses of diverse biomedical multiview datasets, we demonstrate that Coracle consistently outperforms existing non-conformalized methods in both uncertainty estimation and calibration. By equipping published multiview models with valid, interpretable confidence measures, Coracle advances the development of trustworthy, uncertainty-aware statistical tools for biomedical decision support. Coracle is publicly available as an open-source R package at https://github.com/himelmallick/Coracle .
Open AccessConformalized Multiview Learning
S ummary: There is an overwhelmingly large body of literature and numerous algorithms already available on “multimodal AI”, based on different modeling and fusion paradigms. As multimodal AI becomes fully integrated into healthcare applications, the ability to say “I’m not sure” or “I don’t know” when uncertain is a necessary capability for safe clinical deployment, requiring the augmentation of current multimodal AI techniques. Our goal in this work is not to add yet another method to the growing toolbox of multimodal AI, but rather to (i) clarify how existing supervised multimodal AI methods can be understood through the lens of uncertainty quantification, and (ii) develop a unified framework that provides valid and interpretable uncertainty estimates for multimodal AI practitioners. We introduce Coracle, a conformalized framework for multimodal AI that seamlessly adapts to early, late, and intermediate fusion. Coracle provides theoretical marginal confidence guarantees and achieves valid finite-sample coverage without relying on distributional assumptions. Through extensive simulations and analyses of diverse biomedical multiview datasets, we demonstrate that Coracle consistently outperforms existing non-conformalized methods in both uncertainty estimation and calibration. By equipping published multiview models with valid, interpretable confidence measures, Coracle advances the development of trustworthy, uncertainty-aware statistical tools for biomedical decision support. Coracle is publicly available as an open-source R package at https://github.com/himelmallick/Coracle .
- Open Access
Bayesian Cooperative Learning for Multimodal Integration
Summary: Bayesian multimodal models that integrate data from multiple sources, studies, or modalities have garnered considerable attention in recent years. However, existing methods either rely on computationally expensive Markov chain Monte Carlo (MCMC) schemes or adopt variational approaches that forgo principled uncertainty quantification. To address this limitation and cater to practical needs, we abandon the MCMC framework and turn to resampling-based posterior inference for multimodal integration. Our method, Bayesian Cooperative Learning (BayesCOOP), embeds fast maximum a posteriori (MAP) estimation within a Bayesian bootstrap, combining a novel jittered group spike-and-slab prior with an efficient expectation–maximization (E M) coordinate descent algorithm under randomly weighted data perturbations. Averaging posterior summaries (MAP estimates) across bootstrap replicates yields approximate posterior samples that retain Bayesian interpretability while avoiding the computational burden of traditional sampling-based inference. We establish theoretical connections between BayesCOOP’s pseudo-posterior averaging and posterior contraction principles, demonstrating near-optimal posterior consistency under sparsity. Extensive simulation studies and analyses of pregnancy multi-omics datasets demonstrate that BayesCOOP substantially outperforms state-of-the-art early, intermediate, and late fusion approaches in estimation, prediction, and uncertainty assessment. The open-source implementation of BayesCOOP is available at https://github.com/himelmallick/BayesCOOP .
Open AccessBayesian Cooperative Learning for Multimodal Integration
Summary: Bayesian multimodal models that integrate data from multiple sources, studies, or modalities have garnered considerable attention in recent years. However, existing methods either rely on computationally expensive Markov chain Monte Carlo (MCMC) schemes or adopt variational approaches that forgo principled uncertainty quantification. To address this limitation and cater to practical needs, we abandon the MCMC framework and turn to resampling-based posterior inference for multimodal integration. Our method, Bayesian Cooperative Learning (BayesCOOP), embeds fast maximum a posteriori (MAP) estimation within a Bayesian bootstrap, combining a novel jittered group spike-and-slab prior with an efficient expectation–maximization (E M) coordinate descent algorithm under randomly weighted data perturbations. Averaging posterior summaries (MAP estimates) across bootstrap replicates yields approximate posterior samples that retain Bayesian interpretability while avoiding the computational burden of traditional sampling-based inference. We establish theoretical connections between BayesCOOP’s pseudo-posterior averaging and posterior contraction principles, demonstrating near-optimal posterior consistency under sparsity. Extensive simulation studies and analyses of pregnancy multi-omics datasets demonstrate that BayesCOOP substantially outperforms state-of-the-art early, intermediate, and late fusion approaches in estimation, prediction, and uncertainty assessment. The open-source implementation of BayesCOOP is available at https://github.com/himelmallick/BayesCOOP .
- Open Access
Repurposing without Reinventing the Wheel - Ensemble Models for Differential Analysis
Inspired by ensemble models in machine learning, we propose a general framework for aggregating multiple distinct base models to enhance the power of published differential association analysis (DAA) methods. We demonstrate this approach by augmenting popular DAA models with one or more biologically motivated alternatives. This creates an ensemble that bypasses the challenge of selecting an optimal model and instead combines the strengths of complementary statistical models to achieve superior performance. Our proposed ensemble learning approach is platform-agnostic and can augment any existing DAA method, providing a general and flexible framework for various downstream modeling tasks across domains and data types. We performed extensive benchmarking across both simulated and experimental datasets spanning single-cell gene expression, bulk transcriptomics, and microbiome metagenomics, where the ensemble strategy vastly outperformed non-ensemble methods, identified more differential patterns than the competing methods, and displayed good control of false positive and false discovery rates across diversified scenarios. In addition to highlighting a substantial performance boost for state-of-the-art DAA methods, this work has practical implications for mitigating the so-called reproducibility crisis in omics data science. An open-source R package implementing the ensemble strategy is publicly available at https://github.com/himelmallick/DAssemble .
Open AccessRepurposing without Reinventing the Wheel - Ensemble Models for Differential Analysis
Inspired by ensemble models in machine learning, we propose a general framework for aggregating multiple distinct base models to enhance the power of published differential association analysis (DAA) methods. We demonstrate this approach by augmenting popular DAA models with one or more biologically motivated alternatives. This creates an ensemble that bypasses the challenge of selecting an optimal model and instead combines the strengths of complementary statistical models to achieve superior performance. Our proposed ensemble learning approach is platform-agnostic and can augment any existing DAA method, providing a general and flexible framework for various downstream modeling tasks across domains and data types. We performed extensive benchmarking across both simulated and experimental datasets spanning single-cell gene expression, bulk transcriptomics, and microbiome metagenomics, where the ensemble strategy vastly outperformed non-ensemble methods, identified more differential patterns than the competing methods, and displayed good control of false positive and false discovery rates across diversified scenarios. In addition to highlighting a substantial performance boost for state-of-the-art DAA methods, this work has practical implications for mitigating the so-called reproducibility crisis in omics data science. An open-source R package implementing the ensemble strategy is publicly available at https://github.com/himelmallick/DAssemble .
- 20 February 2024
- Published by Springer Nature in Computational Statistics
- Vol. 39 (7) , 3855-3880
Bayesian sequential probability ratio test for vaccine efficacy trials
- 20 February 2024
- Published by Springer Nature in Computational Statistics
- Vol. 39 (7) , 3855-3880
Bayesian sequential probability ratio test for vaccine efficacy trials
- 26 December 2023
- Published by Wiley in Statistics in Medicine
- Vol. 43 (5) , 983-1002
An integrated Bayesian framework for multi‐omics prediction and classification
With the growing commonality of multi‐omics datasets, there is now increasing evidence that integrated omics profiles lead to more efficient discovery of clinically actionable biomarkers that enable better disease outcome prediction and patient stratification. Several methods exist to perform host phenotype prediction from cross‐sectional, single‐omics data modalities but decentralized frameworks that jointly analyze multiple time‐dependent omics data to highlight the integrative and dynamic impact of repeatedly measured biomarkers are currently limited. In this article, we propose a novel Bayesian ensemble method to consolidate prediction by combining information across several longitudinal and cross‐sectional omics data layers. Unlike existing frequentist paradigms, our approach enables uncertainty quantification in prediction as well as interval estimation for a variety of quantities of interest based on posterior summaries. We apply our method to four published multi‐omics datasets and demonstrate that it recapitulates known biology in addition to providing novel insights while also outperforming existing methods in estimation, prediction, and uncertainty quantification. Our open‐source software is publicly available at https://github.com/himelmallick/IntegratedLearner.
- 26 December 2023
- Published by Wiley in Statistics in Medicine
- Vol. 43 (5) , 983-1002
An integrated Bayesian framework for multi‐omics prediction and classification
With the growing commonality of multi‐omics datasets, there is now increasing evidence that integrated omics profiles lead to more efficient discovery of clinically actionable biomarkers that enable better disease outcome prediction and patient stratification. Several methods exist to perform host phenotype prediction from cross‐sectional, single‐omics data modalities but decentralized frameworks that jointly analyze multiple time‐dependent omics data to highlight the integrative and dynamic impact of repeatedly measured biomarkers are currently limited. In this article, we propose a novel Bayesian ensemble method to consolidate prediction by combining information across several longitudinal and cross‐sectional omics data layers. Unlike existing frequentist paradigms, our approach enables uncertainty quantification in prediction as well as interval estimation for a variety of quantities of interest based on posterior summaries. We apply our method to four published multi‐omics datasets and demonstrate that it recapitulates known biology in addition to providing novel insights while also outperforming existing methods in estimation, prediction, and uncertainty quantification. Our open‐source software is publicly available at https://github.com/himelmallick/IntegratedLearner.
- 1 January 2023
- Published by American Psychological Association (APA) in Psychological Services
- Vol. 20 (Suppl 2) , 222-231
Impact of substance use disorders on employment for veterans.
Approximately 1.1 million veterans meet criteria for substance use disorders (SUDs) with 1 in 4 struggling with illicit drugs, 4 in 5 struggling with alcohol use, and 1 in 13 struggling with both. The purpose of this study was to examine the impact of SUDs on closure status (exiting with employment, did not exit with employment) for veterans served in a Department of Veterans Affairs' Veterans Health Administration (VHA) vocational rehabilitation (VR) program. Data (N = 2,620) from a VHA VR program in the Veterans Integrated Service Network 12 network were obtained for the purpose of the present study and consisted of veterans enrolled from 2012 to 2018. Findings showed that veterans without SUDs were more likely to exit with employment, and veterans enrolled in transitional work and community-based employment were more likely to exit with employment compared to those within supported employment (SE). Given that SE helps to serve veterans with the most severe psychological or medical conditions (e.g., active psychosis, schizophrenia, bipolar disorder, spinal cord injury, traumatic brain injury), findings suggest veterans are more successful with less serious mental health conditions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
- 1 January 2023
- Published by American Psychological Association (APA) in Psychological Services
- Vol. 20 (Suppl 2) , 222-231
Impact of substance use disorders on employment for veterans.
Approximately 1.1 million veterans meet criteria for substance use disorders (SUDs) with 1 in 4 struggling with illicit drugs, 4 in 5 struggling with alcohol use, and 1 in 13 struggling with both. The purpose of this study was to examine the impact of SUDs on closure status (exiting with employment, did not exit with employment) for veterans served in a Department of Veterans Affairs' Veterans Health Administration (VHA) vocational rehabilitation (VR) program. Data (N = 2,620) from a VHA VR program in the Veterans Integrated Service Network 12 network were obtained for the purpose of the present study and consisted of veterans enrolled from 2012 to 2018. Findings showed that veterans without SUDs were more likely to exit with employment, and veterans enrolled in transitional work and community-based employment were more likely to exit with employment compared to those within supported employment (SE). Given that SE helps to serve veterans with the most severe psychological or medical conditions (e.g., active psychosis, schizophrenia, bipolar disorder, spinal cord injury, traumatic brain injury), findings suggest veterans are more successful with less serious mental health conditions. (PsycInfo Database Record (c) 2023 APA, all rights reserved).
- 21 November 2022
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 53 (9) , 4352-4362
Unified reciprocal LASSO estimation via least squares approximation
The primary goal of this article is to extend the reciprocal LASSO for applications to binary and survival outcomes. We consider the least squares approximation (LSA) as a solver for the reciprocal LASSO problem. The LSA is a general theoretical framework that includes generalized linear models, Cox regression, and many others as special cases. By applying LSA to reciprocal LASSO regularization, we transfer the original reciprocal LASSO problem into an asymptotically equivalent least squares problem. While the existing literature on reciprocal LASSO has mostly focused on linear models, our algorithm can be easily implemented for general likelihoods, providing a flexible framework for variable selection using reciprocal penalties. To handle the computational burden of implementing the resulting procedure, we employ a scalable stochastic search method called Simplified Shotgun Stochastic Search with Screening (S5), which is easy to implement, without requiring any sophisticated optimization package other than a linear equation solver. We examine the effectiveness of our procedure through Monte Carlo simulations and real data analyses.
- 21 November 2022
- Published by Taylor & Francis in Communications in Statistics - Simulation and Computation
- Vol. 53 (9) , 4352-4362
Unified reciprocal LASSO estimation via least squares approximation
The primary goal of this article is to extend the reciprocal LASSO for applications to binary and survival outcomes. We consider the least squares approximation (LSA) as a solver for the reciprocal LASSO problem. The LSA is a general theoretical framework that includes generalized linear models, Cox regression, and many others as special cases. By applying LSA to reciprocal LASSO regularization, we transfer the original reciprocal LASSO problem into an asymptotically equivalent least squares problem. While the existing literature on reciprocal LASSO has mostly focused on linear models, our algorithm can be easily implemented for general likelihoods, providing a flexible framework for variable selection using reciprocal penalties. To handle the computational burden of implementing the resulting procedure, we employ a scalable stochastic search method called Simplified Shotgun Stochastic Search with Screening (S5), which is easy to implement, without requiring any sophisticated optimization package other than a linear equation solver. We examine the effectiveness of our procedure through Monte Carlo simulations and real data analyses.
- Open Access
An integrated Bayesian framework for multi-omics prediction and classification
With the growing commonality of multi-omics datasets, there is now increasing evidence that integrated omics profiles lead to the more efficient discovery of clinically actionable biomarkers that enable better disease outcome prediction and patient stratification. Several methods exist to perform host phenotype prediction from crosssectional, single-omics data modalities but decentralized frameworks that jointly analyze multiple time-dependent omics data to highlight the integrative and dynamic impact of repeatedly measured biomarkers are currently limited. In this article, we propose a novel Bayesian ensemble method to consolidate prediction by combining information across several longitudinal and cross-sectional omics data layers. Unlike existing frequentist paradigms, our approach enables uncertainty quantification in prediction as well as interval estimation for a variety of quantities of interest based on posterior summaries. We apply our method to four published multi-omics datasets and demonstrate that it recapitulates known biology in addition to providing novel insights while also outperforming existing methods in estimation, prediction, and uncertainty quantification. Our open-source software is publicly available at https://github.com/himelmallick/IntegratedLearner .
Open AccessAn integrated Bayesian framework for multi-omics prediction and classification
With the growing commonality of multi-omics datasets, there is now increasing evidence that integrated omics profiles lead to the more efficient discovery of clinically actionable biomarkers that enable better disease outcome prediction and patient stratification. Several methods exist to perform host phenotype prediction from crosssectional, single-omics data modalities but decentralized frameworks that jointly analyze multiple time-dependent omics data to highlight the integrative and dynamic impact of repeatedly measured biomarkers are currently limited. In this article, we propose a novel Bayesian ensemble method to consolidate prediction by combining information across several longitudinal and cross-sectional omics data layers. Unlike existing frequentist paradigms, our approach enables uncertainty quantification in prediction as well as interval estimation for a variety of quantities of interest based on posterior summaries. We apply our method to four published multi-omics datasets and demonstrate that it recapitulates known biology in addition to providing novel insights while also outperforming existing methods in estimation, prediction, and uncertainty quantification. Our open-source software is publicly available at https://github.com/himelmallick/IntegratedLearner .
- 10 May 2022
- Published by Springer Nature in Sankhya A
- Vol. 85 (1) , 999-1013
Jeffreys Prior for Negative Binomial and Zero Inflated Negative Binomial Distributions
- 10 May 2022
- Published by Springer Nature in Sankhya A
- Vol. 85 (1) , 999-1013
Jeffreys Prior for Negative Binomial and Zero Inflated Negative Binomial Distributions
- 11 November 2021
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 50 (4) , 848-870
A more powerful test for three-arm non-inferiority via risk difference: Frequentist and Bayesian approaches
Necessity for finding improved intervention in many legacy therapeutic areas are of high priority. This has the potential to decrease the expense of medical care and poor outcomes for many patients. Typically, clinical efficacy is the primary evaluating criteria to measure any beneficial effect of a treatment. Albeit, there could be situations when several other factors (e.g. side-effects, cost-burden, less debilitating, less intensive, etc.) which can permit some slightly less efficacious treatment options favorable to a subgroup of patients. This often leads to non-inferiority (NI) testing. NI trials may or may not include a placebo arm due to ethical reasons. However, when included, the resulting three-arm trial is more prudent since it requires less stringent assumptions compared to a two-arm placebo-free trial. In this article, we consider both Frequentist and Bayesian procedures for testing NI in the three-arm trial with binary outcomes when the functional of interest is risk difference. An improved Frequentist approach is proposed first, which is then followed by a Bayesian counterpart. Bayesian methods have a natural advantage in many active-control trials, including NI trial, as it can seamlessly integrate substantial prior information. In addition, we discuss sample size calculation and draw an interesting connection between the two paradigms.
- 11 November 2021
- Published by Taylor & Francis in Journal of Applied Statistics
- Vol. 50 (4) , 848-870
A more powerful test for three-arm non-inferiority via risk difference: Frequentist and Bayesian approaches
Necessity for finding improved intervention in many legacy therapeutic areas are of high priority. This has the potential to decrease the expense of medical care and poor outcomes for many patients. Typically, clinical efficacy is the primary evaluating criteria to measure any beneficial effect of a treatment. Albeit, there could be situations when several other factors (e.g. side-effects, cost-burden, less debilitating, less intensive, etc.) which can permit some slightly less efficacious treatment options favorable to a subgroup of patients. This often leads to non-inferiority (NI) testing. NI trials may or may not include a placebo arm due to ethical reasons. However, when included, the resulting three-arm trial is more prudent since it requires less stringent assumptions compared to a two-arm placebo-free trial. In this article, we consider both Frequentist and Bayesian procedures for testing NI in the three-arm trial with binary outcomes when the functional of interest is risk difference. An improved Frequentist approach is proposed first, which is then followed by a Bayesian counterpart. Bayesian methods have a natural advantage in many active-control trials, including NI trial, as it can seamlessly integrate substantial prior information. In addition, we discuss sample size calculation and draw an interesting connection between the two paradigms.