A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.

High-resolution posture tracking of C. elegans has applications in genetics, neuroscience, and drug screening. While classic methods can reliably track isolated worms on uniform backgrounds, they fail when worms overlap, coil, or move in complex environments. Model-based tracking and deep learning a...

Full description

Saved in:
Bibliographic Details
Main Authors: Weheliye H Weheliye, Javier Rodriguez, Luigi Feriani, Avelino Javer, Virginie Uhlmann, André E X Brown
Format: Article
Language:English
Published: Public Library of Science (PLoS) 2025-08-01
Series:PLoS Computational Biology
Online Access:https://doi.org/10.1371/journal.pcbi.1013345
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1849228253305241600
author Weheliye H Weheliye
Javier Rodriguez
Luigi Feriani
Avelino Javer
Virginie Uhlmann
André E X Brown
author_facet Weheliye H Weheliye
Javier Rodriguez
Luigi Feriani
Avelino Javer
Virginie Uhlmann
André E X Brown
author_sort Weheliye H Weheliye
collection DOAJ
description High-resolution posture tracking of C. elegans has applications in genetics, neuroscience, and drug screening. While classic methods can reliably track isolated worms on uniform backgrounds, they fail when worms overlap, coil, or move in complex environments. Model-based tracking and deep learning approaches have addressed these issues to an extent, but there is still significant room for improvement in tracking crawling worms. Here we train a version of the DeepTangle algorithm developed for swimming worms using a combination of data derived from Tierpsy tracker and hand-annotated data for more difficult cases. DeepTangleCrawl (DTC) outperforms existing methods, reducing failure rates and producing more continuous, gap-free worm trajectories that are less likely to be interrupted by collisions between worms or self-intersecting postures (coils). We show that DTC enables the analysis of previously inaccessible behaviours and increases the signal-to-noise ratio in phenotypic screens, even for data that was specifically collected to be compatible with legacy trackers including low worm density and thin bacterial lawns. DTC broadens the applicability of high-throughput worm imaging to more complex behaviours that involve worm-worm interactions and more naturalistic environments including thicker bacterial lawns.
format Article
id doaj-art-c5159425f3ad4e43bc25f5e0c49fadd3
institution Kabale University
issn 1553-734X
1553-7358
language English
publishDate 2025-08-01
publisher Public Library of Science (PLoS)
record_format Article
series PLoS Computational Biology
spelling doaj-art-c5159425f3ad4e43bc25f5e0c49fadd32025-08-23T05:31:15ZengPublic Library of Science (PLoS)PLoS Computational Biology1553-734X1553-73582025-08-01218e101334510.1371/journal.pcbi.1013345A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.Weheliye H WeheliyeJavier RodriguezLuigi FerianiAvelino JaverVirginie UhlmannAndré E X BrownHigh-resolution posture tracking of C. elegans has applications in genetics, neuroscience, and drug screening. While classic methods can reliably track isolated worms on uniform backgrounds, they fail when worms overlap, coil, or move in complex environments. Model-based tracking and deep learning approaches have addressed these issues to an extent, but there is still significant room for improvement in tracking crawling worms. Here we train a version of the DeepTangle algorithm developed for swimming worms using a combination of data derived from Tierpsy tracker and hand-annotated data for more difficult cases. DeepTangleCrawl (DTC) outperforms existing methods, reducing failure rates and producing more continuous, gap-free worm trajectories that are less likely to be interrupted by collisions between worms or self-intersecting postures (coils). We show that DTC enables the analysis of previously inaccessible behaviours and increases the signal-to-noise ratio in phenotypic screens, even for data that was specifically collected to be compatible with legacy trackers including low worm density and thin bacterial lawns. DTC broadens the applicability of high-throughput worm imaging to more complex behaviours that involve worm-worm interactions and more naturalistic environments including thicker bacterial lawns.https://doi.org/10.1371/journal.pcbi.1013345
spellingShingle Weheliye H Weheliye
Javier Rodriguez
Luigi Feriani
Avelino Javer
Virginie Uhlmann
André E X Brown
A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
PLoS Computational Biology
title A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
title_full A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
title_fullStr A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
title_full_unstemmed A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
title_short A neural network model enables worm tracking in challenging conditions and increases signal-to-noise ratio in phenotypic screens.
title_sort neural network model enables worm tracking in challenging conditions and increases signal to noise ratio in phenotypic screens
url https://doi.org/10.1371/journal.pcbi.1013345
work_keys_str_mv AT weheliyehweheliye aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT javierrodriguez aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT luigiferiani aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT avelinojaver aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT virginieuhlmann aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT andreexbrown aneuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT weheliyehweheliye neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT javierrodriguez neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT luigiferiani neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT avelinojaver neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT virginieuhlmann neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens
AT andreexbrown neuralnetworkmodelenableswormtrackinginchallengingconditionsandincreasessignaltonoiseratioinphenotypicscreens