Publications

Summary

Gillette TA, Ascoli GA (2015) Topological characterization of neuronal arbor morphology via sequence representation: I – Motif analysis. BMC Bioinformatics, 16. Summary | PDF

Gillette TA, Hosseini P, Ascoli GA (2015) Topological characterization of neuronal arbor morphology via sequence representation: II – Global alignmentBMC Bioinformatics, 16. Summary | PDF

Polavaram S, Gillette TA, Parekh R, Ascoli GA (2014) Statistical analysis and data mining of digital reconstructions of dendritic morphologies. Frontiers in Neuroanatomy, 8:138.

Gillette TA, Brown KM, Ascoli GA (2011) The DIADEM Metric: Comparing multiple reconstructions of the same neuronNeuroinformatics, 9:233-45. Summary

Gillette TA, Brown KM, Svoboda K, Liu Y, Ascoli GA (2011) DIADEMchallenge.Org: A compendium of resources fostering the contiuous development of automated neuronal reconstruction. Neuroinformatics, 9:2-3.

Gillette TA, Ascoli GA. Measuring and Modeling Morphology: How Dendrites Take Shape. In Le Novere N. (Ed.), “Computational Systems Neurobiology”, pp. 387-428, Springer (2012). Summary

Theodore C Dumas, Todd A Gillette, Deveroux Ferguson et al. (2010) Anti-glucocorticoid gene therapy reverses the impairing effects of elevated corticosterone on spatial memory, hippocampal neuronal excitability, ...


Gillette TA, Ascoli GA (2015) Topological characterization of neuronal arbor morphology via sequence representation: I – Motif analysis. BMC Bioinformatics, 16.
Summary | PDF

Gillette TA, Hosseini P, Ascoli GA (2015) Topological characterization of neuronal arbor morphology via sequence representation: II – Global alignmentBMC Bioinformatics, 16.
Summary | PDF

Polavaram S, Gillette TA, Parekh R, Ascoli GA (2014) Statistical analysis and data mining of digital reconstructions of dendritic morphologies. Frontiers in Neuroanatomy, 8:138.

Gillette TA, Brown KM, Ascoli GA (2011) The DIADEM Metric: Comparing multiple reconstructions of the same neuronNeuroinformatics, 9:233-45.
Summary

Gillette TA, Brown KM, Svoboda K, Liu Y, Ascoli GA (2011) DIADEMchallenge.Org: A compendium of resources fostering the contiuous development of automated neuronal reconstruction. Neuroinformatics, 9:2-3.

Gillette TA, Ascoli GA. Measuring and Modeling Morphology: How Dendrites Take Shape. In Le Novere N. (Ed.), “Computational Systems Neurobiology”, pp. 387-428, Springer (2012).
Summary

Theodore C Dumas, Todd A Gillette, Deveroux Ferguson et al. (2010) Anti-glucocorticoid gene therapy reverses the impairing effects of elevated corticosterone on spatial memory, hippocampal neuronal excitability, and synaptic plasticity. Journal of Neuroscience 30 (5):1712-1720.
Summary

Todd A Gillette, John J Grefenstette (2009) On Comparing Neuronal Morphologies with the Constrained Tree-edit-distance. In Neuroinformatics 7 (3):191-194.
Summary

Kerry M Brown, Todd A Gillette, Giorgio A Ascoli (2008) Quantifying neuronal size: summing up trees and splitting the branch difference. In Seminars in Cell & Developmental Biology 19 (6):485-493.
Summary

For a copy of any of these publications you may contact me directly or visit the publications request page of the Computational Neuroanatomy Group (part of the CN3 at the Krasnow Institute for Advanced Study).


Summaries

Topological characterization of neuronal arbor morphology via sequence representation: I – motif analysis.

We present a novel representation of neuronal arborizations as sequences of branch points, with each branch encoded based on whether its children branch or terminate. This representation is amenable to a number of different sequence analysis techniques exploited heavily in genetic sequencing. Motif analysis of these neurite sequences shows that axons and dendrites appear more similar to each other than to pyramidal cell apical dendrites when looking at longer and more complex subsequences. These findings offer potential paths of exploration in understanding axon growth principles, and the technique can be used for validating morphological models in greater detail than previously possible. The representation can be easily expanded to include other branch properties like tortuosity and branch angle.

Topological characterization of neuronal arbor morphology via sequence representation: II – global alignment.

We used the sequence representation described in the companion paper for analysis via global sequence alignment. The alignment procedure requires modification from traditional sequence alignment in order to maintain the topological relationships and ensure that morphologies with very different topological structure do achieve a high similarity measure. By clustering the neurites by the alignment similarity (converted into an abstract feature space), we found differences between known neuron class types, including a differentiation between mouse and rat cortical pyramidal axons, which was unexpected. By aligning all co-clustered neurites at one time we were able to extract a consensus for each cluster which shows the global topological features which are responsible for the clusters and which distinguish the neuron classes. We found that cortical perisomatic-targeting interneuron axons have a much more symmetric structure than axons of both dendritic-targeting interneurons as well as pyramidal cells. This may be due to tighter signal timing requirements of perisomatic-targeting interneurons (responsible for network synchrony) and/or effects of more heterogenous and diverse connectivity of the pyramidal and dendritic-targeting interneurons which would increase network complexity. Alignment-derived features turned out to be largely independent of the local motifs found previously.

The DIADEM Metric: Comparing multiple reconstructions of the same neuron.

The article provides background on the challenges in automated reconstruction and describes a measure of reconstruction quality, the DIADEM metric, given an expertly reconstructed gold standard. This measure can additionally be used to gauge the capabilities of reconstruction algorithms. The DIADEM metric is compared to expert judgement, correlating as well to expert judgement and measures of time saved compared to manual reconstruction and editing results of automated reconstruction as those judgements correlate to each other. Ways to adapt the measure to different goals are described, some made possible in the code and others requiring further development.

Measuring and Modeling Morphology: How Dendrites Take Shape.

Dendritic morphology is a complex field related to many other fields including studies of connectivity, electrophysiology, development, and neuronal plasticity, among others. Moreover, a variety of computational methods have been brought to bear to make the most of the growing quality and quantity of data being produced over the past few decades. This book chapter begins with an overview of the basic relationships between neuronal morphology and the various related fields and then delves into several of the major or particularly interesting methods and studies which use dendritic morphology to better understand neuronal development and function.

Anti-glucocorticoid gene therapy reverses the impairing effects of elevated corticosterone on spatial memory, hippocampal neuronal excitability, and synaptic plasticity.

A method for protecting brains, particularly the Hippocampus, from the detrimental effects of excessive stress is explored. The method is a dentate gyrus targeted gene therapeutic injection that produces a molecule which breaks down corticosterone, a glucocorticoid (GC) and the major stress hormone in rodents. A variety of behavioral and electrophysiological effects are seen, particularly a reduction in the GC-induced excitability increase found in dentate gyrus granule cells and CA1 pyramidal cells (synaptically downstream of the granule cells). Synaptic plasticity and memory were also improved by the treatment relative to the high-GC group.

On Comparing Neuronal Morphologies with the Constrained Tree-edit-distance

The Tree Edit Distance (TED) is a measure of how different two trees are from each other, or how many edits it would take to turn one tree into the other. Heumann and Wittum explored the use of an effective heuristic of the TED in classifying neuronal trees. This commentary in the same issue looks at how their approach relates to similar approaches in other fields such as botany, the potential additional applications within neuroscience along with some constraints of the method.

Quantifying neuronal size: summing up trees and splitting the branch difference.

In this part review, part original research article, the main focus is on the ways neuronal morphology has been used to better understand neuronal development and function. Various measurement and modeling techniques are explored, and the new measurement caulescence is introduced. As in the botanical world, some “trees” are very bushy, while others have a clear main trunk. Sometimes that is seen in a literal trunk such as in tall trees or apical dendrites of pyramidal cells in the neocortex. Other times, like in vines or many axons, diameter is irrelevant while the main path can still be seen in the consistent asymmetry along the path. These ideas are presented as a method for determining the main path of a tree structure and for measuring the prominence of that path, or its caulescence. The concept is applied to neuronal trees and compared with associated morphometrics.

Skip to toolbar