Identification of Genes that Influence Sleep & Wake, using a Novel,
High-throughout, Piezoelectric Technology
Bruce O’Hara
(Department of Biology)
Kevin D. Donohue and Hank Dietz
(Department of
Recent advances in science and technology have led
to many new opportunities for studies involving the relationship between
genetic codes and their expressions.
The need for engineering contributions in these explorations was
articulated by Human Genome Project (HGP)
pioneer Charles
DeLisi, “As experimental methods become
increasingly powerful, the mathematical and computational methods of systems
engineering will be essential for converting data to knowledge,…” (Genomes: 15 Years Later,’ Human
Genome News, Vol 11, No. 3-4, July
2001). In that spirit, this
project focuses on developing technology to help determine the genes associated
with sleep behavior. Knowledge
concerning the genetic basis for sleep can provided additional clues for
unlocking the mysteries of sleep
with the potential to enhance the lives of those suffering from sleep disorders
and aid those who must operate in a sleep deprived mode, such as emergency
personnel and soldiers.
|
Dr.
Bruce O’Hara, PI of the project, with single cage unit |
This web site describes the technology we are developing
to perform this task. This work
is being funded in part by a Department of Defense (DoD).DEPSCoR
grant. The project seeks to
identify mice with unusual sleep behaviors. And then through analyzing the
genetic differences between mice with normal and outlying sleep behavior,
identify those genes associated with sleep. The technology for this study requires
monitoring hundreds of mice for days at a time and automatically computing percentage
of time each mouse spent sleeping. |
|
|
The
resulting system to achieve the goal of this study includes:
·
An
array of motion sensors made from Polyvinylidene fluoride (PVDF) material (manufactured by MSI)
·
Cages
to conveniently house the mice with PVDF sensor, associated hardware. and
cabling for interfacing to a PC (See Picture).
·
Sensor
Amplifiers (See Picture)
·
Hardware and
Software interface to acquire and monitor stored data (Developed with data
cards and LabVI
·
Signal
processing to classify stored signal segments into sleep and wake states
(developed using Matlab from Mathworks)
|
An
initial study has just been completed as part of a Masters Thesis by Dharshan
Charith Medonza, Piezoelectric Sensors for Sleep
Detection for Mice, 12/2005, in which he
showed the error in classifying short segments of sensor data (every 4
seconds) a misclassification rate of less than 10% was achievable. |
|
|
|
The
technology described on these pages may be helpful for similar studies that require
monitoring a larger number of animals for long periods of times. Please contact us if there is interest
in collaborations.