Bound2Learn: a machine learning approach for classification of DNA-bound proteins from single-molecule tracking experiments

Abstract

DNA-bound proteins are essential elements for the maintenance, regulation, and use of the genome. The time they spend bound to DNA provides useful information on their stability within protein complexes and insight into the understanding of biological processes. Single-particle tracking allows for direct visualization of protein–DNA kinetics, however, identifying whether a molecule is bound to DNA can be non-trivial. Further complications arise when tracking molecules for extended durations in processes with slow kinetics. We developed a machine learning approach, termed Bound2Learn, using output from a widely used tracking software, to robustly classify tracks in order to accurately estimate residence times. We validated our approach in silico, and in live-cell data from Escherichia coli and Saccharomyces cerevisiae. Our method has the potential for broad utility and is applicable to other organisms.

Following the tracks: How transcription factor binding dynamics control transcription

“…More photostable and brighter organic dyes will allow more accurate tracking of long DNA-binding events, possibly revealing the existence of rare, much longer-lived TF-DNA-bound subpopulations ( 101 ). In addition, advances in analysis software and mathematical modeling will allow more precise characterization of the dynamic properties of different TF subpopulations to provide a deeper understanding of the mechanisms underlying TF-binding dynamics and how they control transcription ( 27 , 32 , 64 , 66 , 102 ).…”

Mars, a molecule archive suite for reproducible analysis and reporting of single-molecule properties from bioimages

“…The discovery of new biological phenomena from single-molecule observations often depends on time-consuming manual classification of individual molecules and behaviors. Machine learning algorithms are now offering the possibility to automate these tasks ( Kapadia et al, 2021 ; Thomsen et al, 2020 ), but their accuracy depends on robust training datasets. The powerful record tagging tools provided with Mars provide the ideal platform for the creation of large training datasets for machine learning based classifiers.…”

Mars, a molecule archive suite for reproducible analysis and reporting of single-molecule properties from bioimages

“…The discovery of new biological phenomena from singlemolecule observations often depends on time-consuming manual classification of individual molecules and behaviors. Machine learning algorithms are now offering the possibility to automate these tasks (Kapadia et al ., 2021; Thomsen et al ., 2020), but their accuracy depends on robust training datasets. The powerful record tagging tools provided with Mars provide the ideal platform for the creation of large training datasets for machine learning based classifiers.…”