Brian Trippe

All Publications

• Calibrating Generative Models to Distributional Constraints Smith, H. D., Diamant, N. L., Trippe, B. L. arXiv. 2025
• MotifBench: A standardized protein design benchmark for motif-scaffolding problems Zheng, Z., Zhang, B., Didi, K., Yang, K. K., Yim, J., Watson, J., Chen, H., Trippe, B. L. arXiv. 2025
• Predicting mutational effects on protein binding from folding energy International Conference on Machine Learning Arthur, D. L., Householder, K., Wu, F., Thrun, S., Garcia, K. C., Trippe, B. L. 2025
• De novo design of protein structure and function with RFdiffusion. Nature Watson, J. L., Juergens, D., Bennett, N. R., Trippe, B. L., Yim, J., Eisenach, H. E., Ahern, W., Borst, A. J., Ragotte, R. J., Milles, L. F., Wicky, B. I., Hanikel, N., Pellock, S. J., Courbet, A., Sheffler, W., Wang, J., Venkatesh, P., Sappington, I., Torres, S. V., Lauko, A., De Bortoli, V., Mathieu, E., Ovchinnikov, S., Barzilay, R., Jaakkola, T. S., DiMaio, F., Baek, M., Baker, D. 2023; 620 (7976): 1089-1100

  Abstract

  There has been considerable recent progress in designing new proteins using deep-learning methods1-9. Despite this progress, a general deep-learning framework for protein design that enables solution of a wide range of design challenges, including de novo binder design and design of higher-order symmetric architectures, has yet to be described. Diffusion models10,11 have had considerable success in image and language generative modelling but limited success when applied to protein modelling, probably due to the complexity of protein backbone geometry and sequence-structure relationships. Here we show that by fine-tuning the RoseTTAFold structure prediction network on protein structure denoising tasks, we obtain a generative model of protein backbones that achieves outstanding performance on unconditional and topology-constrained protein monomer design, protein binder design, symmetric oligomer design, enzyme active site scaffolding and symmetric motif scaffolding for therapeutic and metal-binding protein design. We demonstrate the power and generality of the method, called RoseTTAFold diffusion (RFdiffusion), by experimentally characterizing the structures and functions of hundreds of designed symmetric assemblies, metal-binding proteins and protein binders. The accuracy of RFdiffusion is confirmed by the cryogenic electron microscopy structure of a designed binder in complex with influenza haemagglutinin that is nearly identical to the design model. In a manner analogous to networks that produce images from user-specified inputs, RFdiffusion enables the design of diverse functional proteins from simple molecular specifications.

  View details for DOI 10.1038/s41586-023-06415-8

  View details for PubMedID 37433327

  View details for PubMedCentralID PMC10468394

• Leveraging polygenic enrichments of gene features to predict genes underlying complex traits and diseases. Nature genetics Weeks, E. M., Ulirsch, J. C., Cheng, N. Y., Trippe, B. L., Fine, R. S., Miao, J., Patwardhan, T. A., Kanai, M., Nasser, J., Fulco, C. P., Tashman, K. C., Aguet, F., Li, T., Ordovas-Montanes, J., Smillie, C. S., Biton, M., Shalek, A. K., Ananthakrishnan, A. N., Xavier, R. J., Regev, A., Gupta, R. M., Lage, K., Ardlie, K. G., Hirschhorn, J. N., Lander, E. S., Engreitz, J. M., Finucane, H. K. 2023

  Abstract

  Genome-wide association studies (GWASs) are a valuable tool for understanding the biology of complex human traits and diseases, but associated variants rarely point directly to causal genes. In the present study, we introduce a new method, polygenic priority score (PoPS), that learns trait-relevant gene features, such as cell-type-specific expression, to prioritize genes at GWAS loci. Using a large evaluation set of genes with fine-mapped coding variants, we show that PoPS and the closest gene individually outperform other gene prioritization methods, but observe the best overall performance by combining PoPS with orthogonal methods. Using this combined approach, we prioritize 10,642 unique gene-trait pairs across 113 complex traits and diseases with high precision, finding not only well-established gene-trait relationships but nominating new genes at unresolved loci, such as LGR4 for estimated glomerular filtration rate and CCR7 for deep vein thrombosis. Overall, we demonstrate that PoPS provides a powerful addition to the gene prioritization toolbox.

  View details for DOI 10.1038/s41588-023-01443-6

  View details for PubMedID 37443254

  View details for PubMedCentralID 5501872

• Confidently Comparing Estimates with the c-value JOURNAL OF THE AMERICAN STATISTICAL ASSOCIATION Trippe, B. L., Deshpande, S. K., Broderick, T. 2024; 119 (546): 983-994

  View details for DOI 10.1080/01621459.2022.2153688

  View details for Web of Science ID 000939457600001

• Gaussian Processes at the Helm(holtz): A More Fluid Model for Ocean Currents Berlinghieri, R., Trippe, B. L., Burt, D. R., Giordano, R., Srinivasan, K., Ozgokmen, T., Xia, J., Broderick, T. edited by Krause, A., Brunskill, E., Cho, K., Engelhardt, B., Sabato, S., Scarlett, J. JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2023

  View details for Web of Science ID 001206894302011

• Diffusion probabilistic modeling of protein backbones in 3D for the motif-scaffolding problem International Conference on Learning Representations Trippe, B. L., Yim, J., Tischer, D., Broderick, T., Baker, D., Barzilay, R., Jaakkola, T. 2023
• Practical and Asymptotically Exact Conditional Sampling in Diffusion Models Wu, L., Trippe, B. L., Naesseth, C. A., Blei, D. M., Cunningham, J. P. edited by Oh, A., Neumann, T., Globerson, A., Saenko, K., Hardt, M., Levine, S. NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2023

  View details for Web of Science ID 001227224008039

• SE(3) diffusion model with application to protein backbone generation Yim, J., Trippe, B. L., De Bortoli, V., Mathieu, E., Doucet, A., Barzilay, R., Jaakkola, T. edited by Krause, A., Brunskill, E., Cho, K., Engelhardt, B., Sabato, S., Scarlett, J. JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2023

  View details for Web of Science ID 001372555102002

• Randomized gates eliminate bias in sort-seq assays PROTEIN SCIENCE Trippe, B. L., Huang, B., DeBenedictis, E. A., Coventry, B., Bhattacharya, N., Yang, K. K., Baker, D., Crawford, L. 2022; 31 (9)

  View details for DOI 10.1002/pro.4401

  View details for Web of Science ID 000847526000001

• Many Processors, Little Time: MCMC for Partitions via Optimal Transport Couplings Nguyen, T. D., Trippe, B. L., Broderick, T. edited by Camps-Valls, G., Ruiz, F. J., Valera JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2022

  View details for Web of Science ID 000828072703024

• For high-dimensional hierarchical models, consider exchangeability of effects across covariates instead of across datasets Trippe, B. L., Finucane, H. K., Broderick, T. edited by Ranzato, M., Beygelzimer, A., Dauphin, Y., Liang, P. S., Vaughan, J. W. NEURAL INFORMATION PROCESSING SYSTEMS (NIPS). 2021

  View details for Web of Science ID 000922928201068

• LR-GLM: High-Dimensional Bayesian Inference Using Low-Rank Data Approximations Trippe, B. L., Huggins, J. H., Agrawal, R., Broderick, T. edited by Chaudhuri, K., Salakhutdinov, R. JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2019

  View details for Web of Science ID 000684034306047

• The Kernel Interaction Trick: Fast Bayesian Discovery of Pairwise Interactions in High Dimensions Agrawal, R., Huggins, J. H., Trippe, B., Broderick, T. edited by Chaudhuri, K., Salakhutdinov, R. JMLR-JOURNAL MACHINE LEARNING RESEARCH. 2019

  View details for Web of Science ID 000684034300015

• Inhibition of cell fate repressors secures the differentiation of the touch receptor neurons of Caenorhabditis elegans. Development (Cambridge, England) Zheng, C., Jin, F. Q., Trippe, B. L., Wu, J., Chalfie, M. 2018; 145 (22)

  Abstract

  Terminal differentiation generates the specialized features and functions that allow postmitotic cells to acquire their distinguishing characteristics. This process is thought to be controlled by transcription factors called 'terminal selectors' that directly activate a set of downstream effector genes. In Caenorhabditis elegans, the differentiation of both the mechanosensory touch receptor neurons (TRNs) and the multidendritic nociceptor FLP neurons uses the terminal selectors UNC-86 and MEC-3. The FLP neurons fail to activate TRN genes, however, because a complex of two transcriptional repressors (EGL-44/EGL-46) prevents their expression. Here, we show that the ZEB family transcriptional factor ZAG-1 promotes TRN differentiation not by activating TRN genes but by preventing the expression of EGL-44/EGL-46. As EGL-44/EGL-46 also inhibits the production of ZAG-1, these proteins form a bistable, negative-feedback loop that regulates the choice between the two neuronal fates.

  View details for DOI 10.1242/dev.168096

  View details for PubMedID 30291162

  View details for PubMedCentralID PMC6262790

• Overpruning in variational Bayesian neural networks NeurIPS Workshop on Advances in Approximate Bayesian Inference Trippe, B. L., Turner, R. E. 2017