MJMartin Jankowiak

Martin Jankowiak

*This is an algorithmically generated author record based on publicly available information.

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


Martin Jankowiak has a total of 26 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Inferring selection effects in SARS-CoV-2 with Bayesian Viral Allele Selection

    by Martin Jankowiak, Fritz H. Obermeyer, Jacob E. Lemieux

    The global effort to sequence millions of SARS-CoV-2 genomes has provided an unprecedented view of viral evolution. Characterizing how selection acts on SARS-CoV-2 is critical to developing effective, long-lasting vaccines and other treatments, but the scale and complexity of genomic surveillance data make rigorous analysis challenging. To meet this challenge, we develop Bayesian Viral Allele Selection (BVAS), a principled and scalable probabilistic method for inferring the genetic determinants of differential viral fitness and the relative growth rates of viral lineages, including newly emergent lineages. After demonstrating the accuracy and efficacy of our method through simulation, we apply BVAS to 6.9 million SARS-CoV-2 genomes. We identify numerous mutations that increase fitness, including previously identified mutations in the SARS-CoV-2 Spike and Nucleocapsid proteins, as well as mutations in non-structural proteins whose contribution to fitness is less well characterized. In addition, we extend our baseline model to identify mutations whose fitness exhibits strong dependence on vaccination status as well as pairwise interaction effects, i.e. epistasis. Strikingly, both these analyses point to the pivotal role played by the N501 residue in the Spike protein. Our method, which couples Bayesian variable selection with a diffusion approximation in allele frequency space, lays a foundation for identifying fitness-associated mutations under the assumption that most alleles are neutral.


  • Open Access

    Fast Bayesian Variable Selection in Binomial and Negative Binomial Regression

    by Martin Jankowiak

    Bayesian variable selection is a powerful tool for data analysis, as it offers a principled method for variable selection that accounts for prior information and uncertainty. However, wider adoption of Bayesian variable selection has been hampered by computational challenges, especially in difficult regimes with a large number of covariates or non-conjugate likelihoods. Generalized linear models for count data, which are prevalent in biology, ecology, economics, and beyond, represent an important special case. Here we introduce an efficient MCMC scheme for variable selection in binomial and negative binomial regression that exploits Tempered Gibbs Sampling (Zanella and Roberts, 2019) and that includes logistic regression as a special case. In experiments we demonstrate the effectiveness of our approach, including on cancer data with seventeen thousand covariates.

  • Open Access

    Surrogate Likelihoods for Variational Annealed Importance Sampling

    by Martin Jankowiak, Du Phan

    Variational inference is a powerful paradigm for approximate Bayesian inference with a number of appealing properties, including support for model learning and data subsampling. By contrast MCMC methods like Hamiltonian Monte Carlo do not share these properties but remain attractive since, contrary to parametric methods, MCMC is asymptotically unbiased. For these reasons researchers have sought to combine the strengths of both classes of algorithms, with recent approaches coming closer to realizing this vision in practice. However, supporting data subsampling in these hybrid methods can be a challenge, a shortcoming that we address by introducing a surrogate likelihood that can be learned jointly with other variational parameters. We argue theoretically that the resulting algorithm permits the user to make an intuitive trade-off between inference fidelity and computational cost. In an extensive empirical comparison we show that our method performs well in practice and that it is well-suited for black-box inference in probabilistic programming frameworks.

  • Top Cited Papers
    Open Access

    Analysis of 6.4 million SARS-CoV-2 genomes identifies mutations associated with fitness

    by Fritz Obermeyer, Martin Jankowiak, Nikolaos Barkas, Stephen F. Schaffner, Jesse D. Pyle, Leonid Yurkovetskiy, Matteo Bosso, Daniel J. Park, Mehrtash Babadi, Bronwyn L. MacInnis, et al.

    Repeated emergence of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) variants with increased fitness underscores the value of rapid detection and characterization of new lineages. We have developed PyR 0 , a hierarchical Bayesian multinomial logistic regression model that infers relative prevalence of all viral lineages across geographic regions, detects lineages increasing in prevalence, and identifies mutations relevant to fitness. Applying PyR 0 to all publicly available SARS-CoV-2 genomes, we identify numerous substitutions that increase fitness, including previously identified spike mutations and many nonspike mutations within the nucleocapsid and nonstructural proteins. PyR 0 forecasts growth of new lineages from their mutational profile, ranks the fitness of lineages as new sequences become available, and prioritizes mutations of biological and public health concern for functional characterization.


  • Open Access

    Inferring selection effects in SARS-CoV-2 with Bayesian Viral Allele Selection

    by Martin Jankowiak, Fritz H. Obermeyer, Jacob E. Lemieux

    The global effort to sequence millions of SARS-CoV-2 genomes has provided an unprecedented view of viral evolution. Characterizing how selection acts on SARS-CoV-2 is critical to developing effective, long-lasting vaccines and other treatments, but the scale and complexity of genomic surveillance data make rigorous analysis challenging. To meet this challenge, we develop Bayesian Viral Allele Selection (BVAS), a principled and scalable probabilistic method for inferring the genetic determinants of differential viral fitness and the relative growth rates of viral lineages, including newly emergent lineages. After demonstrating the accuracy and efficacy of our method through simulation, we apply BVAS to 6.9 million SARS-CoV-2 genomes. We identify numerous mutations that increase fitness, including previously identified mutations in the SARS-CoV-2 Spike and Nucleocapsid proteins, as well as mutations in non-structural proteins whose contribution to fitness is less well characterized. In addition, we extend our baseline model to identify mutations whose fitness exhibits strong dependence on vaccination status as well as pairwise interaction effects, i.e. epistasis. Strikingly, both these analyses point to the pivotal role played by the N501 residue in the Spike protein. Our method, which couples Bayesian variable selection with a diffusion approximation in allele frequency space, lays a foundation for identifying fitness-associated mutations under the assumption that most alleles are neutral.


  • Open Access

    Scalable Cross Validation Losses for Gaussian Process Models

    by Martin Jankowiak, Geoff Pleiss

    We introduce a simple and scalable method for training Gaussian process (GP) models that exploits cross-validation and nearest neighbor truncation. To accommodate binary and multi-class classification we leverage P\`olya-Gamma auxiliary variables and variational inference. In an extensive empirical comparison with a number of alternative methods for scalable GP regression and classification, we find that our method offers fast training and excellent predictive performance. We argue that the good predictive performance can be traced to the non-parametric nature of the resulting predictive distributions as well as to the cross-validation loss, which provides robustness against model mis-specification.

  • Open Access

    Analysis of 6.4 million SARS-CoV-2 genomes identifies mutations associated with fitness

    by Fritz Obermeyer, Martin Jankowiak, Nikolaos Barkas, Stephen F. Schaffner, Jesse D. Pyle, Lonya Yurkovetskiy, Matteo Bosso, Daniel J. Park, Mehrtash Babadi, Bronwyn L. MacInnis, et al.

    Repeated emergence of SARS-CoV-2 variants with increased fitness necessitates rapid detection and characterization of new lineages. To address this need, we developed PyR 0 , a hierarchical Bayesian multinomial logistic regression model that infers relative prevalence of all viral lineages across geographic regions, detects lineages increasing in prevalence, and identifies mutations relevant to fitness. Applying PyR 0 to all publicly available SARS-CoV-2 genomes, we identify numerous substitutions that increase fitness, including previously identified spike mutations and many non-spike mutations within the nucleocapsid and nonstructural proteins. PyR 0 forecasts growth of new lineages from their mutational profile, identifies viral lineages of concern as they emerge, and prioritizes mutations of biological and public health concern for functional characterization. One Sentence summary: A Bayesian hierarchical model of all SARS-CoV-2 viral genomes predicts lineage fitness and identifies associated mutations.


  • Open Access

    Deep Sigma Point Processes

    by Martin Jankowiak, Geoff Pleiss, Jacob R. Gardner

    We introduce Deep Sigma Point Processes, a class of parametric models inspired by the compositional structure of Deep Gaussian Processes (DGPs). Deep Sigma Point Processes (DSPPs) retain many of the attractive features of (variational) DGPs, including mini-batch training and predictive uncertainty that is controlled by kernel basis functions. Importantly, since DSPPs admit a simple maximum likelihood inference procedure, the resulting predictive distributions are not degraded by any posterior approximations. In an extensive empirical comparison on univariate and multivariate regression tasks we find that the resulting predictive distributions are significantly better calibrated than those obtained with other probabilistic methods for scalable regression, including variational DGPs--often by as much as a nat per datapoint.

  • Open Access

    Parametric Gaussian Process Regressors

    by Martin Jankowiak, Geoff Pleiss, Jacob R. Gardner

    The combination of inducing point methods with stochastic variational inference has enabled approximate Gaussian Process (GP) inference on large datasets. Unfortunately, the resulting predictive distributions often exhibit substantially underestimated uncertainties. Notably, in the regression case the predictive variance is typically dominated by observation noise, yielding uncertainty estimates that make little use of the input-dependent function uncertainty that makes GP priors attractive. In this work we propose two simple methods for scalable GP regression that address this issue and thus yield substantially improved predictive uncertainties. The first applies variational inference to FITC (Fully Independent Training Conditional; Snelson et.~al.~2006). The second bypasses posterior approximations and instead directly targets the posterior predictive distribution. In an extensive empirical comparison with a number of alternative methods for scalable GP regression, we find that the resulting predictive distributions exhibit significantly better calibrated uncertainties and higher log likelihoods--often by as much as half a nat per datapoint.


  • Open Access

    Fast Matrix Square Roots with Applications to Gaussian Processes and Bayesian Optimization

    by Geoff Pleiss, Martin Jankowiak, David Eriksson, Anil Damle, Jacob R. Gardner

    Matrix square roots and their inverses arise frequently in machine learning, e.g., when sampling from high-dimensional Gaussians $\mathcal{N}(\mathbf 0, \mathbf K)$ or whitening a vector $\mathbf b$ against covariance matrix $\mathbf K$. While existing methods typically require $O(N^3)$ computation, we introduce a highly-efficient quadratic-time algorithm for computing $\mathbf K^{1/2} \mathbf b$, $\mathbf K^{-1/2} \mathbf b$, and their derivatives through matrix-vector multiplication (MVMs). Our method combines Krylov subspace methods with a rational approximation and typically achieves $4$ decimal places of accuracy with fewer than $100$ MVMs. Moreover, the backward pass requires little additional computation. We demonstrate our method's applicability on matrices as large as $50,\!000 \times 50,\!000$ - well beyond traditional methods - with little approximation error. Applying this increased scalability to variational Gaussian processes, Bayesian optimization, and Gibbs sampling results in more powerful models with higher accuracy.