Student or educator? Sign up with your school email and get 6 months of Pro free.
CoChat

The DOI matches its Crossref record; 99 of the 102 references it cites were found in a registry or an index; 17 of the 37 sources its claims cite were checked for support.

Connecting a Connectome to Behavior: An Ensemble of Neuroanatomical Models of C. elegans Klinotaxis

Open access
Izquierdo, Eduardo J., Beer, Randall D. · 2013 · PLoS Computational Biology

Abstract

Increased efforts in the assembly and analysis of connectome data are providing new insights into the principles underlying the connectivity of neural circuits. However, despite these considerable advances in connectomics, neuroanatomical data must be integrated with neurophysiological and behavioral data in order to obtain a complete picture of neural function. Due to its nearly complete wiring diagram and large behavioral repertoire, the nematode worm Caenorhaditis elegans is an ideal organism in which to explore in detail this link between neural connectivity and behavior. In this paper, we develop a neuroanatomically-grounded model of salt klinotaxis, a form of chemotaxis in which changes in orientation are directed towards the source through gradual continual adjustments. We identify a minimal klinotaxis circuit by systematically searching the C. elegans connectome for pathways linking chemosensory neurons to neck motor neurons, and prune the resulting network based on both experimental considerations and several simplifying assumptions. We then use an evolutionary algorithm to find possible values for the unknown electrophsyiological parameters in the network such that the behavioral performance of the entire model is optimized to match that of the animal. Multiple runs of the evolutionary algorithm produce an ensemble of such models. We analyze in some detail the mechanisms by which one of the best evolved circuits operates and characterize the similarities and differences between this mechanism and other solutions in the ensemble. Finally, we propose a series of experiments to determine which of these alternatives the worm may be using.

Overview

CoChat Summary

## TL;DR

The paper builds a neuroanatomically grounded model of *C. elegans* salt klinotaxis by extracting a minimal circuit from the connectome and then using evolutionary algorithms to tune unknown electrophysiological parameters. Across 100 evolutionary runs, 27 high-performing networks produced worm-like klinotaxis, and the authors analyze how information is distributed across AIY, AIZ, and motor neurons.

## Key findings

  • A minimal klinotaxis circuit could be reduced to a fully connected network of 23 neurons (7.61%) and 276 chemical/electrical connections (3.78%) when the path-length constraint was set to 3 and contact threshold to 2 or more contacts.
  • The evolutionary algorithm was run 100 times for 300 generations with populations of 60 individuals; 17 runs failed to produce efficient chemotaxis (CI < 0.5), and 27 networks met the high-performance criterion (CI ≥ 0.75), with average CI = 0.87 on a longer assay.
  • In the successful population, individual AIY cells covered only 0.65 of the sensory range covered by the pair combined on average, while individual AIZ cells covered 0.83; the AIZ gap junction broadened sensitivity more than the AIY gap junction in 24 out of 27 networks.
  • The model reproduced klinotaxis-like behavior: turning bias was an opposite-sign function of bearing, varied linearly with the normal gradient component, and showed a sinusoidal dependence on locomotion phase, matching qualitative patterns seen in real worms.

## Methods

  • The authors mined the *C. elegans* connectome using chemosensory neurons as the root set and neck motor neurons as the target set, progressively constraining the network by path length and contact number to obtain a minimal neuroanatomical klinotaxis circuit.
  • They modeled sensory neurons, interneurons, neck motor neurons, body motion, and the salt gradient environment, using a previous sensorimotor model as a base and adding interneuronal pathways plus electrical synapses.
  • Model parameters were optimized with a genetic algorithm: 60-individual populations, 300 generations, 100 independent runs, fitness measured as a chemotaxis index across 50 assays starting 4.5 cm from the gradient peak.

## Limitations

  • The paper states that the connectome data are not 100% complete; connectivity is partially missing for 39 of 302 neurons, including 21 of 75 motor neurons.
  • The authors explicitly note that the model is not the ultimate model and is a starting point; it relies on simplifying assumptions, incomplete electrophysiological knowledge, and a minimal circuit rather than the full biological network.
  • Evident from the text: the model uses idealized body mechanics, a simplified oscillatory representation, and several untested assumptions about synapse type and strength; this means the findings are constrained by modeling choices rather than direct physiological measurement.

Model-written summary. Check it against the paper before relying on it.

Integrity signals

Registry

Crossref · 10.1371/journal.pcbi.1002890

Cites

99 of 99 references verified0 citation errors · 0 not checkable

Statistics

Nothing to recompute.

Figures

Figure 1 Searching the connectome for the klinotaxis network. A, Maximal network. All paths between chemosensory neurons (white disks) and neck motor neurons (black disks). Interneurons shown in gray. Chemical synapses shown as black lines. Directionality not shown. Gap junctions shown in red. Line thickness represents the relative number of chemical synapses or gap junctions between the two neurons. B, C, Number of neurons and number of chemical synapse and gap junctions in the network as a function of path length. Points represent the number of total neurons and contacts in the network. Dashed lines represent the total number of neurons and synapses and junctions in the connectome.
Figure 1 Searching the connectome for the klinotaxis network. A, Maximal network. All paths between chemosensory neurons (white disks) and neck motor neurons (black disks). Interneurons shown in gray. Chemical synapses shown as black lines. Directionality not shown. Gap junctions shown in red. Line thickness represents the relative number of chemical synapses or gap junctions between the two neurons. B, C, Number of neurons and number of chemical synapse and gap junctions in the network as a function of path length. Points represent the number of total neurons and contacts in the network. Dashed lines represent the total number of neurons and synapses and junctions in the connectome.CC BY · PubMed Central
Figure 2 Neuroanatomy of the minimal klinotaxis circuit. White disks, chemosensory neurons; gray disks, interneurons; dark gray disks, motor neurons. Black connections represent chemical synapses. Red connections represent electrical gap junctions. The strength and polarity of the connections are still unknown. We show the strength and polarity from the best evolved network. Arrowheads, excitatory connections; filled circles, inhibitory connections. Line thickness shows relative connection strength. Color of the disk's border represents the sign of the bias term θ (black, positive; gray, negative). Thickness of the disk's border represents the magnitude of the bias. All parameters were symmetrical across the dorso-ventral midline.
Figure 2 Neuroanatomy of the minimal klinotaxis circuit. White disks, chemosensory neurons; gray disks, interneurons; dark gray disks, motor neurons. Black connections represent chemical synapses. Red connections represent electrical gap junctions. The strength and polarity of the connections are still unknown. We show the strength and polarity from the best evolved network. Arrowheads, excitatory connections; filled circles, inhibitory connections. Line thickness shows relative connection strength. Color of the disk's border represents the sign of the bias term θ (black, positive; gray, negative). Thickness of the disk's border represents the magnitude of the bias. All parameters were symmetrical across the dorso-ventral midline.CC BY · PubMed Central
Figure 3 Behavioral trace. A representative model worm was placed at 10 random orientations 4.5 cm from the gradient peak and allowed to move for 300 sec. The gradient is Gaussian shaped. Trace color represents time. Inset: Enlargement of a gradual turn.
Figure 3 Behavioral trace. A representative model worm was placed at 10 random orientations 4.5 cm from the gradient peak and allowed to move for 300 sec. The gradient is Gaussian shaped. Trace color represents time. Inset: Enlargement of a gradual turn.CC BY · PubMed Central
Figure 4 Klinotaxis analysis. A, Average turning bias vs. bearing. B, Average turning bias vs. the component of the gradient in the normal direction. C, Average turning bias vs. the component of the gradient in the translational direction. Black points represent all of the data averaged into bins according to the translational gradient. Gray points show the data separated between positive and negative turning bias. The two red points highlight the significance of the difference between the turning bias when moving down the gradient versus when moving up the gradient at the same angle. Error bars are standard error of the mean.
Figure 4 Klinotaxis analysis. A, Average turning bias vs. bearing. B, Average turning bias vs. the component of the gradient in the normal direction. C, Average turning bias vs. the component of the gradient in the translational direction. Black points represent all of the data averaged into bins according to the translational gradient. Gray points show the data separated between positive and negative turning bias. The two red points highlight the significance of the difference between the turning bias when moving down the gradient versus when moving up the gradient at the same angle. Error bars are standard error of the mean.CC BY · PubMed Central
Figure 5 Phase sensitivity of orientation responses. A, Response to upsteps. B, Response to downsteps. Plots show turning bias vs. the phase of locomotion at which the concentration step occurred. Each point represents the average across all networks; error bars are standard error of the mean. Shades of gray indicate the magnitude of the concentration step (0.005 black, 0.00333 dark gray, and 0.00166 light gray). Positive and negative values of turning bias result, respectively, in counterclockwise and clockwise rotations of the translational vector. Abbreviations: VS, ventral head sweep; DS, dorsal head sweep (shaded region).
Figure 5 Phase sensitivity of orientation responses. A, Response to upsteps. B, Response to downsteps. Plots show turning bias vs. the phase of locomotion at which the concentration step occurred. Each point represents the average across all networks; error bars are standard error of the mean. Shades of gray indicate the magnitude of the concentration step (0.005 black, 0.00333 dark gray, and 0.00166 light gray). Positive and negative values of turning bias result, respectively, in counterclockwise and clockwise rotations of the translational vector. Abbreviations: VS, ventral head sweep; DS, dorsal head sweep (shaded region).CC BY · PubMed Central
Figure 6 Neuron output traces in response to step changes in concentration in the best evolved network. A, Chemosensory neuron output, ASE. B, First layer interneuron output, AIY. C, Second layer interneuron output, AIZ. D and E, neck motor neuron output (D, SMBD; E, SMBV). Time is shown on the x-axis. Neuron output is shown on the y-axis. Each trace is color-coded according to the magnitude and sign of the step of the concentration step. The black trace represents the neuron output without input. The neuron output from left cells is shown on the left; the neuron output from right cells is shown on the right.
Figure 6 Neuron output traces in response to step changes in concentration in the best evolved network. A, Chemosensory neuron output, ASE. B, First layer interneuron output, AIY. C, Second layer interneuron output, AIZ. D and E, neck motor neuron output (D, SMBD; E, SMBV). Time is shown on the x-axis. Neuron output is shown on the y-axis. Each trace is color-coded according to the magnitude and sign of the step of the concentration step. The black trace represents the neuron output without input. The neuron output from left cells is shown on the left; the neuron output from right cells is shown on the right.CC BY · PubMed Central
Figure 7 Dynamics of the neck motor neurons. Input-output diagrams for the left (A) and right (B) pair of dorsal and ventral motor neurons. Gray trace, steady-state input-output (SSIO) curve when the head sweep oscillation is absent. Red trace, instantaneous input-output relation when the head sweep oscillation is present. Arrows show the effects of output of the indicated chemosensory neuron on motor neuron input. Shaded areas show the range of oscillation. For each of the SSIO curves, there are dorsal and ventral motor neurons moving out of phase over the red trajectory due to the out of phase input from the oscillatory component. When the dorsal motor neuron is at point a in the curve, the ventral motor neuron is at point d, and vice versa. When the dorsal motor neuron is at point b in the curve, the ventral motor neuron is at point c, and vice versa. For the dorsal/ventral SMBL pair (A), an increase/decrease in chemical concentration would decrease/increase the output of the neuron in d, but not of the neuron in a; decreasing/increasing the difference between the two left neck motor neurons. For the dorsal/ventral SMBR pair (B), a and d represent the opposite regions: a neuron at a is more sensitive to changes in input than the other neuron at d; but the result is the same: an increase/decrease in chemical concentration decreases/increases the difference between the two neck right motor neurons. To different degrees, the same is the case for other points along the curve.
Figure 7 Dynamics of the neck motor neurons. Input-output diagrams for the left (A) and right (B) pair of dorsal and ventral motor neurons. Gray trace, steady-state input-output (SSIO) curve when the head sweep oscillation is absent. Red trace, instantaneous input-output relation when the head sweep oscillation is present. Arrows show the effects of output of the indicated chemosensory neuron on motor neuron input. Shaded areas show the range of oscillation. For each of the SSIO curves, there are dorsal and ventral motor neurons moving out of phase over the red trajectory due to the out of phase input from the oscillatory component. When the dorsal motor neuron is at point a in the curve, the ventral motor neuron is at point d, and vice versa. When the dorsal motor neuron is at point b in the curve, the ventral motor neuron is at point c, and vice versa. For the dorsal/ventral SMBL pair (A), an increase/decrease in chemical concentration would decrease/increase the output of the neuron in d, but not of the neuron in a; decreasing/increasing the difference between the two left neck motor neurons. For the dorsal/ventral SMBR pair (B), a and d represent the opposite regions: a neuron at a is more sensitive to changes in input than the other neuron at d; but the result is the same: an increase/decrease in chemical concentration decreases/increases the difference between the two neck right motor neurons. To different degrees, the same is the case for other points along the curve.CC BY · PubMed Central
Figure 8 Range of sensitivity to changes in concentration. A, Interneuron class AIY. B, Interneuron class AIZ. Left neuron shown in blue; right neurons shown in red. Flat regions of the curve correspond to areas of no-sensitivity to input over that region. The slope of the curve denotes the degree of sensitivity to changes around that region of the input. For example, AIYR is sensitive to large positive changes in concentration only, whereas AIYL is sensitive to changes in concentration around the midline. Dashed lines show the sensitivity when the gap junction between those two neurons is blocked. In AIY, blocking the gap junction does not affect the sensitivity of its left and right neurons. In AIZ, blocking the gap junction results in substantial differences in the range of sensitivity. C, D, Sensitivity in the left and right pair of motor neurons as a function of the phase of locomotion, respectively. The dorsal motor neuron is shown in blue-green-yellow shades; the ventral motor neuron is shown in gray scale. Similar to the interneurons, the motor neurons have a selective range of sensory input over which they are sensitive. Unlike the interneurons, their sensitivity changes as a function of the worm's sinusoidal movement. Because of the out of phase oscillation between dorsal and ventral motor neurons, which neuron is more sensitive to a certain input swaps back and forth between them.
Figure 8 Range of sensitivity to changes in concentration. A, Interneuron class AIY. B, Interneuron class AIZ. Left neuron shown in blue; right neurons shown in red. Flat regions of the curve correspond to areas of no-sensitivity to input over that region. The slope of the curve denotes the degree of sensitivity to changes around that region of the input. For example, AIYR is sensitive to large positive changes in concentration only, whereas AIYL is sensitive to changes in concentration around the midline. Dashed lines show the sensitivity when the gap junction between those two neurons is blocked. In AIY, blocking the gap junction does not affect the sensitivity of its left and right neurons. In AIZ, blocking the gap junction results in substantial differences in the range of sensitivity. C, D, Sensitivity in the left and right pair of motor neurons as a function of the phase of locomotion, respectively. The dorsal motor neuron is shown in blue-green-yellow shades; the ventral motor neuron is shown in gray scale. Similar to the interneurons, the motor neurons have a selective range of sensory input over which they are sensitive. Unlike the interneurons, their sensitivity changes as a function of the worm's sinusoidal movement. Because of the out of phase oscillation between dorsal and ventral motor neurons, which neuron is more sensitive to a certain input swaps back and forth between them.CC BY · PubMed Central
Figure 9 Orientation responses elicited by stepwise changes in concentration. A, Response to upsteps. B, Response to downsteps. 1, Behavioral traces for four points in the locomotion phase. The black trace shows the trajectory in the absence of a concentration step. The colored traces show the trajectories obtained in response to concentration steps placed at the locations indicated by dots of matching color. The four locations chosen cover a complete cycle of locomotion (dashed box). Vectors indicating the direction of translation (dashed lines) are shown for no concentration step (u), and steps at locations a and c (ua, and uc). Abbreviations: DS, dorsal head sweep; VS, ventral head sweep (shaded region). 2, Generalization of the turning bias as a function of the timing within the locomotion phase where the step is introduced. The example traces from A1 and B1 are represented by labeled disks in A2 and B2, respectively.
Figure 9 Orientation responses elicited by stepwise changes in concentration. A, Response to upsteps. B, Response to downsteps. 1, Behavioral traces for four points in the locomotion phase. The black trace shows the trajectory in the absence of a concentration step. The colored traces show the trajectories obtained in response to concentration steps placed at the locations indicated by dots of matching color. The four locations chosen cover a complete cycle of locomotion (dashed box). Vectors indicating the direction of translation (dashed lines) are shown for no concentration step (u), and steps at locations a and c (ua, and uc). Abbreviations: DS, dorsal head sweep; VS, ventral head sweep (shaded region). 2, Generalization of the turning bias as a function of the timing within the locomotion phase where the step is introduced. The example traces from A1 and B1 are represented by labeled disks in A2 and B2, respectively.CC BY · PubMed Central
Figure 10 Interneuron types based on their response to step changes in concentration. Type A: insensitive. Type B: sensitive only to either upsteps or downsteps. Type C: ambiguously sensitive to both upsteps and downsteps. Type D: unambiguously sensitive to upsteps and downsteps. Figures in column 1 show neuron output trace over time, colored according to the different steps in concentration received. The sign and magnitude of the step in concentration is given by the color bar. The black trace represents the output of the neuron without any input. Column 2 shows their corresponding sensitivity plot. Average output of the neuron (relative to the average output of the neuron in the absence of any input) as a function of step changes in concentration of different sign and magnitude over the range of sensory stimuli. Flat regions of the curve correspond to areas of no-sensitivity to input over that region. The slope of the curve denotes the degree of sensitivity to changes around that region of the input.
Figure 10 Interneuron types based on their response to step changes in concentration. Type A: insensitive. Type B: sensitive only to either upsteps or downsteps. Type C: ambiguously sensitive to both upsteps and downsteps. Type D: unambiguously sensitive to upsteps and downsteps. Figures in column 1 show neuron output trace over time, colored according to the different steps in concentration received. The sign and magnitude of the step in concentration is given by the color bar. The black trace represents the output of the neuron without any input. Column 2 shows their corresponding sensitivity plot. Average output of the neuron (relative to the average output of the neuron in the absence of any input) as a function of step changes in concentration of different sign and magnitude over the range of sensory stimuli. Flat regions of the curve correspond to areas of no-sensitivity to input over that region. The slope of the curve denotes the degree of sensitivity to changes around that region of the input.CC BY · PubMed Central
Figure 11 Effects of simulated neuron ablations on klinotaxis performance. Neuron activation set to resting potential (such that neuron output is 0) during assay. The middle bar inside the box is the statistical median. The outer edges of the box represent the 0.25 and 0.75 quantiles. The lines that extend out of the box span the full dataset.
Figure 11 Effects of simulated neuron ablations on klinotaxis performance. Neuron activation set to resting potential (such that neuron output is 0) during assay. The middle bar inside the box is the statistical median. The outer edges of the box represent the 0.25 and 0.75 quantiles. The lines that extend out of the box span the full dataset.CC BY · PubMed Central
Figure 12 Interneuron comparison. A, Unique information about the sensory input in the interneurons. The coverage of a cell measures the range of the sensory stimuli over which it is sensitive to changes. When the left and right cells specialize over a different range of possible sensory stimuli, their combined coverage can be greater than their average independent coverage. The x-axis represents the gain in coverage when the left and right AIY cells are considered jointly from when each cells is considered independently. The y-axis represents the same information for the AIZ pair. B, Role of gap junctions in the interneurons. Effects of simulated blocked gap junction on coverage of the interneurons over the range of step changes in concentration for AIY and AIZ cells. The x-axis represents the loss in coverage when the AIY gap junction is blocked, compared to when it is functioning normally. The y-axis represents the same information for the AIZ pair. Each point represents a network from the successful population.
Figure 12 Interneuron comparison. A, Unique information about the sensory input in the interneurons. The coverage of a cell measures the range of the sensory stimuli over which it is sensitive to changes. When the left and right cells specialize over a different range of possible sensory stimuli, their combined coverage can be greater than their average independent coverage. The x-axis represents the gain in coverage when the left and right AIY cells are considered jointly from when each cells is considered independently. The y-axis represents the same information for the AIZ pair. B, Role of gap junctions in the interneurons. Effects of simulated blocked gap junction on coverage of the interneurons over the range of step changes in concentration for AIY and AIZ cells. The x-axis represents the loss in coverage when the AIY gap junction is blocked, compared to when it is functioning normally. The y-axis represents the same information for the AIZ pair. Each point represents a network from the successful population.CC BY · PubMed Central
Figure 13 Neuroanatomical network connecting ASE to SMB, constrained to paths of length 3, but without a contact threshold. White disks, chemosensory neurons; gray disks, interneurons; dark gray disks, motor neurons. (Note: ADFR is a chemosensory neuron, but it is not shown in white because it is not in the original target set. Similarly, RMGL is a motor neuron, but it is not shown in black because it is not in the motor target set). Black connections represent chemical synapses; red connections represent electrical gap junctions.
Figure 13 Neuroanatomical network connecting ASE to SMB, constrained to paths of length 3, but without a contact threshold. White disks, chemosensory neurons; gray disks, interneurons; dark gray disks, motor neurons. (Note: ADFR is a chemosensory neuron, but it is not shown in white because it is not in the original target set. Similarly, RMGL is a motor neuron, but it is not shown in black because it is not in the motor target set). Black connections represent chemical synapses; red connections represent electrical gap junctions.CC BY · PubMed Central
Figure 14 Model worm. A, Idealized body. The model worm was represented as a point (x, y) located at the center of the border between idealized head and neck regions of the model. The quantity μ is the angle between the instantaneous velocity vector v and the positive x axis. The dorsal (gray) and ventral (black) motor neuron pairs receive out of phase oscillatory input from the body. B, Changes in orientation. Between time steps i−1 and i, the orientation of the velocity vector changes by the turning angle φ. The gray arc is the model worm's path. C, Terminology. Orientation vectors used in the analysis of sinusoidal locomotion. Undulations occur in the x–y plane. The white circles represent the start and end of one locomotion cycle. Solid section of the trajectory represents ventral head sweeps. Dashed section of the trajectory represents dorsal head sweeps.
Figure 14 Model worm. A, Idealized body. The model worm was represented as a point (x, y) located at the center of the border between idealized head and neck regions of the model. The quantity μ is the angle between the instantaneous velocity vector v and the positive x axis. The dorsal (gray) and ventral (black) motor neuron pairs receive out of phase oscillatory input from the body. B, Changes in orientation. Between time steps i−1 and i, the orientation of the velocity vector changes by the turning angle φ. The gray arc is the model worm's path. C, Terminology. Orientation vectors used in the analysis of sinusoidal locomotion. Undulations occur in the x–y plane. The white circles represent the start and end of one locomotion cycle. Solid section of the trajectory represents ventral head sweeps. Dashed section of the trajectory represents dorsal head sweeps.CC BY · PubMed Central

Verification

Claims the paper makes

13 Valid · 5 Not valid · 15 Not checked · 150 Own results and interpretations

109 more claims

Create a free account to see all 183 claims and to verify the 2 references still open.

See all claims — free account
  • Borderline: one of the tests for a substantial claim is unclear. 3
  • It may state two separate claims as one. 57
  • Probably not a substantial claim. 14

Create a free account to see these claims.

Citations

Cites

Show all 102 references

Research works printed but not linked: 1

  1. Schnitzer MJ, Block SM, Berg HC, Purcell EM (1990). Strategies for chemotaxis. In: Armitage JP, Lackie JM, eds. Biology of the Chemotactic Response. Cambridge: Cambridge University Press. pp 15–34.

Cited by

See all 90

Metadata

Authors
Izquierdo, E. J., & Beer, R. D.
Affiliations
Indiana University Bloomington (US)
Corresponding author
Izquierdo, Eduardo J.
Venue
PLoS Computational Biology
Published
2013-02-07
Volume
9
Issue
2
Pages
e1002890
ISSN
  • 1553-734X (ISSN-L)
  • 1553-7358
OpenAlex
W2034131443
Cited by
84
References
102
Citations by year
2026: 5 · 2025: 6 · 2024: 6 · 2023: 3 · 2022: 6 · 2021: 6 · 2020: 4 · 2019: 6 · 2018: 10 · 2017: 7 · 2016: 9 · 2015: 9 · 2014: 6 · 2013: 1
FWCI
4.08
Type
journal article
Language
en
Open access
Locations
  • PLoS Computational Biology (publishedVersion, cc-by)
  • Europe PMC (PubMed Central) (publishedVersion, cc-by)
  • PubMed Central (publishedVersion, cc-by)
  • PubMed (publishedVersion)
  • DOAJ (DOAJ: Directory of Open Access Journals) (submittedVersion, cc-by-sa)
  • Figshare (submittedVersion, cc-by)
  • IUScholarWorks (Indiana University) (submittedVersion, cc-by)
Funders
National Science Foundation (1216739, 0916409)
Topics
  • Life Sciences › Neuroscience › Cellular and Molecular Neuroscience › Neurobiology and Insect Physiology Research
  • Life Sciences › Neuroscience › Cognitive Neuroscience › Neural dynamics and brain function
  • Social Sciences › Education › Undergraduate Neuroscience Education and Research
Keywords
Caenorhabditis elegans, klinotaxis, connectome, neural circuit model, neuroanatomical model, chemosensory neurons, network pruning, evolutionary algorithms, ensemble learning, behavioral optimization
MeSH terms
Algorithms, Animals, Behavior, Animal, Chemotaxis, Models, Neurological, Motor Neurons, Sensory Receptor Cells, Caenorhabditis elegans, Computational Biology, Connectome

About these checks

Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.

The claim verdicts this report counts are experimental: their accuracy is not yet measured.

A language model drew the claims from the paper on 2026-10-10, under version 3 of the claim definition (model deepseek-v4.1-flash, prompt extract-v3).

A second model labelled each claim on 2026-10-10 (model jev-1.13-20260917, definition revision gold-guideline-v0.2).

Cited sources are checked with a language model’s help (gpt-6-sol).

Statistics and arithmetic are recomputed, not judged.

Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.

What counts as a claim ›How we measure confidence ›How we check this ›

Are you an author, or do you see a problem with this page? Request a review

How to cite

Cite CoChat for the checks on this page, and the paper for its own findings.

This page
CoChat. (2026). Citation check: Connecting a Connectome to Behavior: An Ensemble of Neuroanatomical Models of C. elegans Klinotaxis. https://app.cochat.ai/papers/10.1371/journal.pcbi.1002890
The paper
Izquierdo, E. J., & Beer, R. D. (2013). Connecting a Connectome to Behavior: An Ensemble of Neuroanatomical Models of C. elegans Klinotaxis. PLoS Computational Biology. https://doi.org/10.1371/journal.pcbi.1002890

Data sources

Bibliographic data: Semantic Scholar, OpenAlex, Crossref.

Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).

Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.