University of Pittsburgh | Pitt's Mathematics Department        
     
 
Tim Stilson
 

        

February 2004

1 2 3 4 5 6 7
8 9 10 11 12 13 14
15 16 17 Chapter Meeting 18 19 20 21
22 23 24 Chapter Meeting 25 26 27 28
29            

March 2004

 

home
calendar
officers
by-laws
links
photos
contact
how to join

 

 

problems with the site?
webmaster

 

17th -- Chapter Meeting
               Speaker: Stephanie Hoogendoorn
               Title: "Computational Modeling of the Thalamus in the Rat
               Whisker Barrel Sensory System"
               Where: 704 Thackeray
               Time: 3:00 PM (Tuesday)
                Refreshments will be served immediately afterward in 705.
                Abstract: Computational Modeling of Synaptic Dynamics underlying the temporal responses of
                          thalamic barreloid neurons to periodic whisker deflections.

                           Previous experimental studies suggest that frequency-dependent properties of thalamic barreloid
                           neurons to periodic whisker deflections reflect the effects of synaptic depression, low-threshold
                           calcium T-current, and local inhibition of neurons in the ventral posterior medial nucleus (VPM) by
                           the thalamic reticular nucleus (nRt). We incorporate synaptic depression and T-current (Sohall,
                           Huguenard 1998) into a quadratic integrate-and-fire model in order to simulate a network of VPM
                           and nRt neurons. Input consists of spike trains obtained previously from VPM-projecting neurons
                           in brainstem nucleus principalis (PrV). Stimuli consisted of periodic sinusoidal or pulsatilewhisker
                           deflections over the frequency range 1-40 Hz. Model neurons, like their biological counterparts,
                           display adaptation, with responses of both VPM and nRt being larger during the first vs subsequent
                           cycles. Also, simulated and real nRT cells display more sustained and less temporally distinct
                           discharges over each deflection cycle. Analysis of model neurons indicate that synaptic depression
                            of the PrV to VPM pathway can account for decremented VPM and nRt responses to later cycles
                            in the stimulus train. We also find that T-current enhanced nRt spiking, and hence synaptic
                            inhibition. T-current has a greater effect at lower frequencies in both VPM and nRt, and further,
                            the effect is more pronounced for (high velocity) pulsatile deflections than for (lower velocity)
                           sinusoidal movements. T-current acts by prolonging and temporally smoothing nRT responses.
                           Prolonged nRT firing can also be produced by simulating long-lasting synapses at VPM to nRt
                           connections. Together, PrV-VPm synaptic depression and low-pass nRT characteristics enhance
                            the temporal modulation of VPm responses by temporally focusing activity evoked by whisker
                            movement while suppressing spiking during inter-deflection periods. (grant number NIMH 61372.)

24th -- Chapter Meeting
               Speaker: Glen Whitehead
               Title: "Graph Theory Applied to Chemistry"
               Where: 704 Thackeray
                Time: 3:00 PM (Tuesday)
                Abstract: In this presentation, two applications of graph theory to chemistry are discussed.  The
                          first application is boiling point models of alkanes.  The second application is melting point models
                          of a special class of alkanes.

                          Burch, Wakefield and Whitehead developed boiling point models for alkane isomers containing up
                          to and including 12 carbon atoms.  These models are based on the following indices: molecular
                          weight, number of carbon atoms, Hosoya index, Wiener number, Wiener Path numbers, methyl index
                          and ethyl index.  Nonlinear regression techniques were used to determine coefficients and exponents
                          of indices in our models.

                          Burch and Whitehead developed melting point models for a special class of alkanes.  This class
                          contains alkanes having 10 to 20 carbon atoms and only one methyl group.  This class of alkanes
                          are of special interest to petroleum engineers interested in manufacturing synthetic diesel fuel.
                          These models are based on the following indices: number of carbon atoms, Wiener Path numbers,
                           mean Wiener index, and the methyl locant index.  Nonlinear regression techniques were used to
                          determine coefficients and exponents of the indices in our model.