Timing in Biological Systems Lou Scheffer Howard Hughes Medical - PowerPoint PPT Presentation

Timing in Biological Systems Lou Scheffer Howard Hughes Medical Institute http://www.tiempo-secure.com/technology/asynchronous-design-technology/

Nervous systems and electronics � A lot in common � Basic operation is electrical � Multiple inputs + nonlinearity � Both have circuits where timing is critical � Tools for studying timing � Experimental � Theoretical

Biology has cool tricks � Grows (no $6B factory) � Resilient � Learns Credits: Lynn Riddiford, Wojciech Maly, Rex Kerr

Even politicians recognize this is a good problem to work on!

Trying to understand the brain � Hot topic – many methods are being used � Structural – reverse engineering � Genetic � Behavioral � Electrophysiology � Imaging � Lineage � Combinations of these techniques Philosophy, Washinton and Lee

History – Basic Science � Caton saw electrical activity in animal brains, 1875 � Berger produced first EEGs of humans in 1924 � Hodgkin-Huxley Nobel-winning model in 1953 � Established basic model of how a neuron works From Nobelprize.org

Neuron operation � Internal operation � Chemical � Electrical � A voltage controlled current source � Timing is critical to the main neuron operation – “action potential” or spike

Action potential result of time constants � http://hyperphysics.phy-astr.gsu.edu/hbase/biology/actpot.html

Also ‘gap’ junctions � Direct connection between cell interiors � Like a resistor (diode) � Same sign � Gain < 1 � Very fast � Not used much in mammalian brains � Used in insect brains � Used in the retina and visual system of mammals Wikipedia Gap cell junction en.svg 1 Nov 2009

Example circuits where timing is critical � Motion detection � Sound localization � Energy savings & synchronization � Learning http://www.theguardian.com/us-news/2014/oct/29/paraglider-dell-schanze- charged-with-harassing-an-owl-in-utah

Motion detection depends on delays L1 L1 � t � t � � Mi1 Mi1 + + + + + + + + Tm3 Σ Σ _/ _/ T4 T4 + - + - Σ Σ Credit: Dmitri Chklovskii, Janelia, HHMI

Hearing and localization depend on precise delays Carr, C. E., and M. Konishi. "A circuit for detection of interaural time differences in the brain stem of the barn owl." The Journal of Neuroscience 10.10 (1990): 3227-3246.

Nature builds balanced trees Carr, C. E., and M. Konishi. "A circuit for detection of interaural time differences in the brain stem of the barn owl." The Journal of Neuroscience 10.10 (1990): 3227-3246.

Different neurons respond differently to the same inputs ������������������������������������������������������ ���������������������������������

Energy savings or synchronization � Data from locust (not in all insects) Perez-Orive, Javier, et al. "Oscillations and sparsening of odor representations in the mushroom body." Science 297.5580 (2002): 359-365.

Spike Timing Dependent Plasticity ������������� ����������������!�������������������������� ����������������������������������"#� � ������������������������ �����$������� %��&�'��(�����)��)��&�*���*$������+������,�-,���##� ��.- ���������������������������������������

Spike Timing Dependent Plasticity � IBM trying to implement this with memristors Adjust M weights in one cycle Multi-purpose Neuro-architecture with Memristors Idongesit Ebong, Durgesh Deshpande, Yalcin Yilmaz, and Pinaki Mazumder

Repeater insertion in biology � We know there must be repeaters � RC of neurons gives 1 ms for a 1 cm length � If strictly electrical, goes like L^2 � 10 sec for a 1 meter nerve � Not observed, so potential is regenerated � Propagates like a wavefront (constant speed) � A main success of Hodges/Huxley model � As in EE, two regimes � Small insects, neurons are isopotential and propagation delays are << gates � Large animals (mammals) transport delays dominate

Long nerves are different � Bigger diameter helps, but only as sqrt(d) � Myelinated nerves reduce capacitance Axon Dendrite terminal Node of Ranvier Soma Schwann cell Myelin sheath Nucleus Wikipedia: Neuron Hand-tuned.svg 1 Nov 2009

How does this help? � Myelin is a good insulator � Isolates sections of line � Less C and less leakage � Needs to reach threshold by the next node � Gives factor of a few better speed � Very similar to repeater insertion in VLSI � Myelin is white, which is why the brain is grey matter (no myelin) interconnected by white matter

Statistical timing in biology � Every step in bio timing is statistical � Small number (usually 1?) of vesicles released during an action potential � Each is small (~5000 molecules) so a ~70 molecule std deviation � This triggers a few (1-10? Ion channels) � That relax statistically….. � Combines effects of statistical timing and noise

Ion channels are fundamentally statistical � Open channel always the same size � Controlled by a Markov-like process THE PHYSICAL BASIS OF ION CHANNEL KINETICS: THE IMPORTANCE OF DYNAMICS, Liebovitch and Krekora http://www.ccs.fau.edu/~liebovitch/dyn.html

Biological systems stable to perturbations in timing � Cold blooded animals work over a range of 10- 40 degrees C. � Reaction rates vary considerably over this range. � But animal behavior largely unchanged � Not well understood how this works

Tools for studying timing � In one way, similar to studying timing in ICs � Most methods cannot measure signals without affecting them � One big problem – no test chips � Some readout techniques too slow � Some have promise of sufficient speed, but not yet � Fast-enough methods have other limitations.

Gold standard: single cell recording � Excellent resolution � Can see sub-threshold � Both input and output � Drawbacks � Invasive � Hard to connect to desired cell � Short life (1/2 hour) � Not easy to parallelize � Tough in behaving animals

Electrophysiology (continued) � Animals such as mice and rats can wear headgear � Now working for flies using ‘virtual reality’ Michael Reiser, janelia, HHMI

Electrical readout by electrode array � Electrodes go near cells, not in them � Readout by capacitive coupling � Each electrode reads many cells (needs spike sorting) � Timing is good, but not cell ID � Spikes only (no subthreshold) ������������������������������������������������������������ ������������������������������������� � ������ ���� � �������������� � ���������������� � ���������������� � ��������������� � ������� �!������� ������� BYU Brown.edu

Spike sorting � All nearby neurons couple to the same probe – data is ambiguous � Spikes of different sizes/speeds probably come from different neurons � Uses classification algorithms; called ‘spike sorting’ � Can potentially record from hundreds of neurons � But we don’t see this many - The ‘dark matter’ problem…. Wikipedia: Spike clusters.png 1 Nov 2009

Extracellular recording (many cells) � IMTEK probes, from CMOS technology 40 µm www.pb.izm.fhg.de 500 µm Source: K. Seidl, et al., JMEMS 2011, vol. 20, pp. 1439 -1448

Consortium for active probes � Readout of hundreds of channels Replace head with LPFs, A/Ds, muxes

Imaging � Use genetic techniques to add indicators that glow when cells are active � Can look at many cells in parallel � Can be too many cells – need genetic subsets � Disadvantages � Limited to surface of brain, except for a few transparent animals, such as zebrafish larva � Voltage resolution is not great � Time constants too long � But better indicators are continually developed

Zebrafish brain imaging Misha Ahrens, Philipp Keller, Janelia, HHMI

Perturbing biological timing � Affect timing of operating neurons � Turn specific neurons on and off � Equivalent of stuck-at faults � Current and voltage injection work well, but are difficult and tedious