Many important datasets in physics, chemistry, and biology consist of noisy sequences of images of multiple moving overlapping particles. In many cases, the observed particles are indistinguishable, leading to unavoidable uncertainty about nearby particles' identities. Exact Bayesian inference is intractable in this setting, and previous approximate Bayesian methods scale poorly. Non-Bayesian approaches that output a single "best" estimate of the particle tracks (thus discarding important uncertainty information) are therefore dominant in practice. Here we propose a flexible and scalable amortized approach for Bayesian inference on this task. We introduce a novel neural network method to approximate the (intractable) filter-backward-sample-forward algorithm for Bayesian inference in this setting. By varying the simulated training data for the network, we can perform inference on a wide variety of data types. This approach is therefore highly flexible and improves on the state of the art in terms of accuracy; provides uncertainty estimates about the particle locations and identities; and has a test runtime that scales linearly as a function of the data length and number of particles, thus enabling Bayesian inference in arbitrarily large particle tracking datasets.

Download data

• Downloaded 939 times
• Download rankings, all-time:
  • Site-wide: 13,548 out of 94,912
  • In bioinformatics: 2,116 out of 8,837
• Year to date:
  • Site-wide: 41,185 out of 94,912
• Since beginning of last month:
  • Site-wide: 54,784 out of 94,912

Altmetric data

Downloads over time

Distribution of downloads per paper, site-wide