Fly Brain

This project is a collaboration with Dr. Gregory Jefferis from the Department of Zoology at Cambridge University (Cambridge, UK).


Olfactory neurons from multiple Drosophila individuals before nonrigid registration (image courtesy of G. Jefferis).

This research aims to understand the transformation of information that occurs in neurons in the brain during the progression from sensation to perception and eventually action. We study this process in the olfactory system of the fruitfly, Drosophila. The project goal is to integrate high-resolution anatomical information about the wiring diagram of the relevant neural circuits with electrophysiological recordings from genetically identified neurons within the brain of the awake behaving fly. Electrophysiology is laborious and technically very challenging but high-resolution neuroanatomy can significantly accelerate analysis of circuit function by predicting which neurons in the circuit are likely to be connected.


Olfactory neurons from multiple Drosophila individuals after nonrigid registration (image courtesy of G. Jefferis).

In constructing a neuroanatomical map one of the critical issues is mapping data obtained from many individual sample brains into a standardised reference co-ordinate system. The same consideration is evident in fruitfly neuroanatomy or mapping of human brains - for example as part of functional magnetic resonance imaging (fMRI) studies of brain activity. Just like human brains, every fly brain is a little different in shape as well as size, so nonrigid registration methods are critical to allow heterogeneous deformation across the sample brain.

The figures on this page show neurons from 30 individual flys. Each color corresponds to a different elementary smell (6 classes). There are five neurons for each class. After nonrigid registration, not only do the neurons cluster very nicely, but so do the colors, i.e., the elementary smells. Note that the registration algorithm works on the original, unprocessed micrscopy images and has no access to the neuron models.

  1. T. Rohlfing, F. Schaupp, D. Haddad, R. Brandt, A. Haase, R. Menzel, and C. R. Maurer, Jr., ``Unwarping Confocal Microscopy Images of Bee Brains by Nonrigid Registration to a Magnetic Resonance Microscopy Image,'' Journal of Biomedical Optics, in press.
  2. T. Rohlfing and C. R. Maurer, Jr., ``Nonrigid image registration in shared-memory multiprocessor environments with application to brains, breasts, and bees,'' IEEE Transactions on Information Technology in Biomedicine, vol. 7, no. 1, pp. 16-25, 2003.

URL: http://www.stanford.edu/~rohlfing/research/fly/index.html
Last updated September 08 2011 16:24:19.
Torsten Rohlfing, Ph.D., torsten@synapse.sri.com
SRI International, Neuroscience Program
333 Ravenswood Avenue, Menlo Park, CA 94025-3493, USA