Fall 2018 Schedule

08/30/2018Anthony Aportela / Yasemin Ozkan AydinFenton / Bhamla
09/06/2018Olga ShishkovHu Lab
09/13/2018Thomas SpencerHu Lab
09/20/2018Gabi SteinbachYunker
09/27/2018Bahnisikha DuttaGoldman
10/04/2019Noah DeTalWiesenfeld
10/11/2018David RyooJC Gumbart
10/18/2018Guanlin Li Weitz group
10/25/2018Megan MatthewsSponberg
11/01/2018Alireza ZamaniPeter Yunker
11/08/2018Michael RyanHarold Kim
11/15/2018Shlomi CohenCurtis
11/22/2018Thanksgiving Break
11/29/2018Conner HerndonFenton

Fall 2018 Abstracts


Anthony Aportela / Yasemin Ozkan Aydin

Genetic approach to cardiac modeling (Anthony)

Cardiac Models allow us to better understand the dynamics that occur within a heart. Heart related death is the number one killer in the United States. With models we can better understand the things that harm us in an effort to prevent and treat them. However, the heart is a chaotic system, small changes in parameters can lead to vastly distinct solutions. Finding the right parameters takes a combination of time, experience, and luck, especially for models with many tens of parameters. To expedite this process it is useful to take inspiration from biology and use a genetic algorithm. This talk goes through the theory behind creating a genetic algorithm as well as it’s successes and failures when used to fit real world data

Dynamics of a blob (Yasemin)

Organisms across all length scales (from cells to humans) cluster and forms large social groups for evolutionary advantages. In some cases, aggregates exhibit and enable new functionalities: floating on water (fire ants), nest-building (bees) and mobbing predators (birds). In this talk, we describe a recent discovery of aggregation in flexible living organisms and discuss emergent mechanics, dynamics and collective behavior.


Olga Shishkov

Fly larvae feed by forming a flowing fountain

Black solder fly larvae are edible maggots that are raised by startups all over the world as a source of sustainable protein. A larva competes with its thousands of neighbors to eat twice its body weight per day in decomposing organic waste. We investigate how the collective motion of an aggregation of larvae “pumps” larvae towards a piece of food by considering the feeding behaviors of larvae from individuals to groups of 60,000. We perform time-lapse photography and particle image velocimetry analysis of top and bottom side views of larvae in glass dishes. Around food, larvae from a fountain with their bodies where larvae crawl towards food through the middle of the fountain and fall down the sides once they are done eating. This distributes food between the individuals in the fountain, rather than only allowing a select few larvae to eat.


Thomas Spencer

Sniffing Scaling Study for Superior Sensing

Mammals such as dogs are known for their keen sense of smell and have been relied upon for their ability to find odor sources. A key component to the mammalian sense of smell is the dynamic sniff cycle. We find the rate at which mammals sniff scales at approximately the same rate as their maximum possible sniff frequency. We rationalize this trend due to the limits of their respiratory anatomy and physiology. Lungs of all mammals are constrained to approximately the same pressure whereas the geometry of the system increases with body size. This scaling argument and other literature suggests that mammals sniff as quickly as possible. Conversely, we find through oscillatory wind tunnel experiments and computational simulations that lower sniffing frequencies provide better odor collection in straight, rectangular channels. We proceed from rectangular channels to investigating the effect of biological nasal cavity shapes helps to mitigate odor collection. We apply insights gleaned from our biological and experimental results to design an electronic nose pre-concentrator for improved chemical sensing.


Gabi Steinbach

The Various Views on Killing – or: Antagonistic Interactions in Bacterial Biofilms

Biofilms constitute a relevant part of the microbiome of living organisms. Typically, a vast diversity of different microbial species exists inside a biofilm. This poses a series of challenges as individuals must avoid predation and compete for resources, space, and survival. In response, microbial species have evolved a variety of cooperative and competitive strategies. One potential mechanism for antagonistic killing is the Type VI secretion system (T6SS), which is present in a wide range of Gram-negative species. The T6SS enables bacteria to onject fatal toxins into other bacteria as well as eukaryotic cells. here, I will present studies on biofilms consisting of two antagonistic V. cholerae strains, as they provide an experimentally controllable and practically relevant system of mutual killers. From bacterial assays, the antagonistic behavior of T6SS-active V. cholerae strains has been proven previously by counting the survival rate of bacteria after exposure to an T6SS-active strain. More detailed information on the activity in the biofilm has been gained from microscopy. It shows that such an antagonistic one-on-one interaction causes an initially well-mixed culture to phase separate, providing protection by number. However, in experiments the typical size of clonal patches stops changing much earlier than expected from numerical results, which assume on-contact killing and replication. Here, I will present results on the biological and physical processes at the interface between mutual killing strains of V. cholerae with T6SS. I discuss the relevance of these findings for how clonal patches do – and do not – change, and derive consequences for the role of T6SS in dense biofilms.


Bahnisikha Dutta

Behaviorally Organizing and Buzzing Robots (BOB Bots)

In recent years, collective behavior has become a widely researched topic. Social insects like fire ants and bees, birds, fish schools, etc. provide good experimentation platforms for probing such collective behavior problems such as self-assembly, construction and transport. Alongside biological experimentation, computational modelling has helped understand a few principles governing the emergence of intelligent swarm behaviors from really simple local agents. A relatively new technique to approach these questions is to develop robo- physical swarm platforms for experimentation. The talk is going to introduce one such swarm of robots (BOB Bots) that we designed to help us with two interesting collective problems: a) jamming in active matter collectives and b) self-assembly, organization and transport in low memory or memoryless systems. BOB Bots house a vibrational motor in a 3d printed case and has rudimentary sensing and computation capabilities. These Bots interact with the environment and other BOB bots at a very local level based on contact and light sensing. The stochastic locomotion of each robot with no external control and minimal sensing manifests itself into interesting emergent phenomenon.