Spring 2019

Departmental Seminars

 

January 7, 2019
10:30 AM
Room 1011, EB1
Dr. Michael Hoffmann
California Institute of Technology 
 Ollis Lecture

With support from the Bill and Melinda Gates Foundation, we have developed, tested, and implemented transportable reactor systems and public toilet systems that are designed for the onsite treatment of domestic wastewater with the total recycle of treated water as flushing water.  Our research and development efforts are a major component of the Gates Foundation “Reinvent the Toilet Challenge (RTTC)“.  After pre-treatment with a sequential anaerobic/aerobic baffled bioreactor, the effluent is processed sequentially through semiconductor electrochemical arrays where the COD and microbial loads are reduced to below US EPA reuse standards.  Special mini-reactors are used to convert the treated wastewater to handwashing and drinking water standards.   The treated black water is recycled back into flushing water reservoirs without discharge to the surrounding environment.   Human wastewater can be clarified with the elimination of suspended particles along with >95% reduction in chemical oxygen demand (COD), and a total elimination of fecal coliforms, E. coli, viruses, and total coliforms.  Enteric organism disinfection is achieved for bacteria and viruses via anodic reactive chlorine generation from in situ chloride coupled with cathodic reduction of water to form hydrogen. Improvement of the performance and durability of the core semiconductor anodes along with materials modifications to lower their production costs ongoing. Third-generation prototypes are undergoing field-testing in locations that lack conventional urban infrastructure for wastewater discharge and treatment; the packaged treatment systems can operate without an external source of electricity or fresh water.  Two Caltech-China joint-venture companies, Eco-San and Entrustech, have been established in Yixing, Jiangsu Province to manufacture solar units for the developing world and to provide associated biochemical and electrochemical reactor systems. Additional industrial collaborations have been established in India with ERAM Scientific and with the Kohler Company (USA/India) for production of units to be used in urban and peri-urban environments in India. At the current time, larger-scale systems (> 1000 users/day) are being assembled for use in South Africa, Southern China, and the Philippines.  In late 2017, President Xi Jinping launched the “Chinese Toilet Revolution“ to provide improve public toilets throughout China. Caltech/Eco-San will play a major role in this initiative.

January 14, 2019
10:30 AM
Room 1011, EB1
Dr. Dennis Hess
Georgia Institute of Technology

Engineers typically work in teams where a leader oversees and guides process/product research, development, and direction.  When technically-trained individuals undertake a leadership role, frustration is a frequent outcome, despite technical competency and good intentions.  This situation largely occurs because our technical training has developed within us a mindset wherein we approach problems in a particular way that is based solely on logic, facts, and data.  People are not facts or data, and in many cases, their approach to problems does not appear logical.  Sudden awareness of such behavior is disconcerting to technically-trained individuals who are now in a position of leadership.  The leader must recognize that decisions, even by engineers or scientists, may be driven by biases, priorities, emotions, and previous experiences; the complex interplay of personal attitudes with technical issues has been referred to as sociotechnical problems.  This seminar will discuss reasons why engineers and scientists often find adaptation into leadership roles frustrating and discouraging.  Approaches that facilitate the transition from a purely technical mindset to a technical leader mindset will be described.

February 4, 2019
10:30 AM
Room 1011, EB1
Dr. Abraham Lenhoff
University of Delaware

Protein solutions are ubiquitous in bioprocessing, structural biology and in nature, making it important to understand and predict their physical and thermodynamic properties, including formation of dense phases such as precipitates, crystals, gels and aggregates.  The anisotropic shape and chemical character of protein molecules, on which extensive information down to the atomic level is available from X-ray crystallography, add considerable complexity, especially in describing intermolecular interactions and phase behavior – the molecular sociology.  In particular, the statistical mechanics of anisotropy, including the contributions of strongly attractive interactions guided by the same mechanisms that give rise to biomolecular recognition, can lead to counterintuitive consequences.  This presentation will explore these consequences, with a focus on the structure and evolution of amorphous dense phases of proteins, such as precipitates and gels.

February 11, 2019
10:30 AM
Room 1011, EB1
Dr. Anant Paravastu
Georgia Institute of Technology

The β-sheet is a common tertiary structure for polypeptide chains.  The structure is characterized by extended molecular conformations and inter-chain hydrogen bonding.    Aggregation into nanostructures composed of thousands of molecules can be driven through inter-molecular β-sheet assembly.  This aggregation can be an initiating or propagating event in “protein misfolding” diseases (e.g., Alzheimer’s), producing amyloid fibrils that can infect through self-replication.  The β-sheet can also be the basis for rational design of peptide biomaterials and assemblies of peptide-mimetic molecules (e.g., peptoids).

My research group specializes in solid-state NMR spectroscopy, which makes it possible to precisely assess molecular conformations and inter-molecular packing within β-sheet and similar assemblies.  Many modern solid-state NMR techniques were developed to probe Alzheimer’s amyloid-β fibrils, which I will use as a foundation for discussion.  I will present new structural constraints on a small (oligomeric) aggregate of the same peptide and propose an explanation of why this alternative assembly pathway does not produce aggregate sizes above ~30 molecules.  With insights from the Alzheimer’s amyloid-β, I will then show how we have tested molecular structures formed by rationally designed peptides.  I will show that, while successful designs exist for β-sheet-assembling peptides, unanticipated molecular-level behavior is readily observable and the ability to control assembly pathway is an important frontier.   I will further show that peptides that were not designed to assemble into β-sheets may do so under application-relevant conditions. For peptide analogs, I will show surprising molecular-level similarities to and departures from assembly behaviors of peptides.  In summary, I hope to convince you that molecular structural measurements provide important feedback to the molecular design process for peptide, protein, and peptide analog assembly.

February 18, 2019
10:30 AM
Room 1011, EB1
GSA Panel

The panelists are:

–       Dr. Daniel Armstrong (ABB)

–       Dr. Zach Mundy (BASF)

–       Dr. Brittany Mertens (Biogen)

–       Dr. Mark Schulte (Novozymes)

Often students transitioning from academic research to industrial research may still be unsure of what PhD’s do in industry and how to apply. Four recent CBE PhD alumni from different industries will give their perspective on being a PhD in industry, how they originally obtained a position, and what you can do to prepare. After a brief introduction by our panelists on their education and career trajectory, there will be plenty of time for questions from the audience.

February 25, 2019
10:30 AM
Room 1011, EB1
Dr. Chang Lu
Virginia Tech

Epigenome dictates turning on and off genes during normal development and diseases, forming another layer of regulation on top of gene sequence. Epigenome is cell-type-specific and highly dynamic over the course of disease and treatment thus offers a treasure trove of information for precision medicine. However, there is a giant gap between the number of cells that can be derived from patient samples and millions of cells per assay required by conventional epigenomic assays.

In this seminar, I will discuss the role of microfluidics in conducting genome-wide epigenetic analysis using scarce samples derived from mice and patients. I will argue that microfluidics uniquely facilitates multi-step molecular biology manipulation required by epigenomic assays and interface between the assays and next-generation sequencing. I will describe the microfluidic technologies developed in my lab for profiling histone modifications and DNA methylation. These technologies work in the 30-200 cells per assay range and offer data quality comparable to those of conventional assays. I will also discuss the biological insights we generated into cancer development and brain functions.

March 4, 2019
10:30 AM
Room 1011, EB1
Dr. James van Deventer
Tufts University

Nature uses just 20 amino acid building blocks to generate the proteins that support life. However, nature also frequently augments the limited chemistries of these amino acids in order to accommodate an expanded range of molecular recognition, catalysis, and other complex tasks. In the laboratory, can we leverage more chemical functionality within proteins to solve problems nature has never even considered? This seminar will describe our progress in answering this question, keeping in mind our long-term goal of evolving novel, clinically relevant proteins. Our key advance is in combining the use of yeast display (high throughput protein engineering) with noncanonical amino acid incorporation technology. This approach enables the construction, evaluation, and screening of protein conjugates on the yeast surface using chemistries not normally found in proteins. We have demonstrated all of these capabilities within therapeutically relevant antibody structures. Current work with this platform is exploring the hypothesis that introducing enzyme active site-targeting groups into antibodies will lead to new classes of potent, specific enzyme inhibitors not accessible by current drug discovery methods. In addition, we have found that utilizing noncanonical amino acids with yeast display enables quantitative measurements for characterizing and evolving eukaryotic genetic code manipulation systems. This has already resulted in an expanded range of genetically encodable noncanonical amino acids in yeast, and it raises the possibility of evolving the eukaryotic translation apparatus to accomplish nativelike protein translation with altered genetic codes.

March 11, 2019
10:30 AM
Room 1011, EB1
Dr. Ehssan Nazockdast
UNC Chapel Hill

The cytoskeleton is an active assembly of macromolecular filaments and molecular motors immersed in the cytoplasmic fluid and is involved in vital cellular processes such as cell division and migration. Understanding the mechanics of the cytoskeleton is key to understanding the physics of living systems. To this end, we have developed a highly efficient, large-scale computational platform for dynamic simulation of Stokesian flexible (de)polymerizing fibrous assemblies, their fluid-structure interactions and their emergent cytoplasmic flows. In an integrated computational and experimental study, we combine our computational tool with laser ablation and fluid flow experiments to study the mitotic spindle positioning during cell division, which is crucial for chromosome segregation and asymmetric cell division and involves the interaction of microtubule assemblies with motor-proteins and subcellular organelles. Our findings strongly argue that proper positioning is primarily achieved by the action of motor-proteins bound to the cell boundary. Next, we combine the data from the first full 3D tomographic reconstructions mitotic spindle with modeling and large-scale simulations to choose between different proposed mechanical models for microtubules and chromosomes interactions. Most remarkably, our simulations show that one model can quantitatively reproduce the shapes (curvature) of microtubules measured from tomography results, while others give significantly different predictions.

Fluid dynamics is not a word that appears regularly in cell biology. There has been very little interface between these two important fields of science in the past. I will show how fluid dynamics can be used as a unique diagnostic tool in studying the active forcing mechanisms in positioning of the mitotic spindle during cell division. Yet the conclusions drawn from studying fluid dynamics are quite generic and can used in studying mechanics of other cytoskeletal assemblies.

March 18, 2019
10:30 AM
Room 1011, EB1
Dr. Pablo Debenedetti
Princeton University
Gubbins Lecture

The thermodynamic and kinetic behavior of water in nano-scale confinement plays an important role in biophysical phenomena such as hydrophobically-driven self-assembly. Using advanced sampling techniques, we investigate computationally the rate, mechanism, and energetics of evaporation transitions induced by hydrophobic confinement. We find a pronounced sensitivity of the evaporation kinetics to the substrate’s mechanical properties: a single order of magnitude reduction in the material’s modulus causes the evaporation rate to increase by nine orders of magnitude. Theoretical and numerical analysis of the underlying thermodynamics likewise suggests that it may be possible to sensitively control the relative stability of the vapor and liquid phases by tuning substrate flexibility. These findings may have implications for the function of membrane-bound protein assemblies

March 19, 2019
3:00 PM
Room 135, BTEC
Dr. Pablo Debenedetti
Princeton University
Gubbins Lecture

Water affects every aspect of our lives, from agriculture to climate, and from health to geopolitics. It is a key participant in the physical and chemical processes that sustain life as we know it. Its ubiquity and importance notwithstanding, there remain major open questions about water’s physical properties, which are anomalous by comparison to those of most other liquids. Examples include the fact that the liquid, if sufficiently cold, expands when cooled and becomes less viscous when compressed. Water’s oddities become more pronounced at low temperatures, especially in the supercooled regime, where the liquid is metastable with respect to crystallization. After introducing some of supercooled water’s remarkable properties and their practical implications, I will review some the hypotheses that have been proposed to explain experimental observations. Computer simulations have played an important role in research on supercooled water, in large part because they are not subject to the limitations that make experimental probing of deeply metastable states so challenging. I will illustrate the advantages and limitations of computational investigations of supercooled water, focusing on the intriguing possibility of the existence of a liquid-liquid phase transition. The recent resolution of a long-standing debate on this topic underscores the importance of openness and transparency in research.

March 27, 2019
10:30 AM
Room 1011, EB1
Dr. James Spivey
Louisiana State University

The direct activation and conversion of methane are among the most challenging reactions in chemistry. One approach is based on “superacids”, which were first demonstrated by George Olah in his Nobel Prize research. In principle, the reaction is simple—only one reactant (methane) and only hydrogen and hydrocarbon products. However, a practical process has not been demonstrated. Among other constraints, an industrial process would likely require a solid superacid. Using research at LSU and Univ. Cardiff on gas-phase Al-Br superacid, solid superacids have based shown to be active, essentially a solid analog of the active sites known to be in the  in the gas-phase. Continuing work has focused on developing a high-temperature solid superacid to address the thermodynamic limitations associated with Al-Br catalysts.

April 1, 2019
10:30 AM
Room 1011, EB1
Dr. Fernando Escobedo
Cornell University

As meso-scale building blocks, oligomers, polymers, and nanoparticles can be tailored in ways that atomic or small-molecule building blocks cannot. Recent progress in synthesis and fabrication methods allow the creation of multi-block oligomers and nanoparticles that vary not only in size and chemical composition but also in shape, rigidity, branching topology, and spatial functionalization. A key challenge that such boundless possibilities present to modelers is the ability to predict the assembling patterns of novel building blocks, and thus potentially identify phases with desirable structures and physical, optical, electronic, catalytic or mechanical properties for emerging applications.

I will describe first our efforts to optimize the formation of different types of colloidal alloys, which can be seen as the analog of strategies that have already developed to make useful salts or doped solids from inorganic elements or alloys and intermetallic compounds from metals. The goal is to advance general principles and approaches to design the inter-species interactions between nanoparticles that optimize the formation of either substitutionally disordered alloys or substitutionally ordered alloys. The work focuses on binary mixtures consisting of nanoparticle components whose interactions can be characterized by asymmetries in entropic and energetic characteristics. We have formulated variational principles for enhancing co-assembly behavior with the target type of substitutional order and tested those principles by application to mixtures containing components of diverse size and shape (including polyhedral) and selective interactions that mimic the hybridization of complementary short DNA strands grafted to the nanoparticle surfaces. Some of our specific predictions are consistent with results of nanoparticle alloys already realized.

I will also describe our work on the phase behavior of multiblock oligomers, i.e., molecules consisting of several chemical block types, focusing on cases where a rigid core is one of the constituent blocks. Our interest centers on architectures that can create complex ordered mesophases that combine solid-like domains interspersed with liquid-like domains, which can be used, e.g., as dual electron/ionic conducting materials. Our results are filling some of the gaps in the rich phase behavior that has been mapped experimentally.

Throughout the talk, I will touch on some of the newer methodological variants used to simulate free energies of the systems of interest and how questions related to the optimization of kinetic behavior (beyond thermodynamic stability) could be addressed.

April 8, 2019
10:30 AM
Room 1011, EB1
Dr. Tom Truskett
The University of Texas at Austin

Nanometer-scale, colloidally-stable particles suspended in a fluid can be driven to assemble into a wide variety of different structures depending on the control parameters of the system and the nature of the effective interparticle interactions.  In many cases, the relevant interactions are tunable via external fields, physical or chemical modification of the particle surfaces, or changes in the composition of the suspending solvent. In this talk, we discuss some of the theoretical challenges associated with the inverse design of interactions for assembly into a targeted structure, the detection of such a transition, and the opportunities that new machine learning based simulation approaches provide for addressing both.

April 15, 2019
10:30 AM
Room 1011, EB1
Dr. Joan Brennecke
The University of Texas at Austin

Ionic liquids (ILs) present intriguing possibilities for removal of carbon dioxide from a wide variety of different gas mixtures, including post-combustion flue gas, pre-combustion gases, air, and raw natural gas streams.  Even by physical absorption, many ILs provide sufficient selectivity over N2, O2, CH4 and other gases.  However, when CO2 partial pressures are low, the incorporation of functional groups to chemically react with the CO2 can dramatically increase capacity, while maintaining or even enhancing selectivity.  We will demonstrate several major advances in the development of ILs for CO2 capture applications.  First, we will show how the reaction stoichiometry can be doubled over conventional aqueous amine solutions to reach one mole of CO2 per mole of IL by incorporating the amine on the anion.  Second, we will show how we have been able to virtually eliminate any viscosity increase upon complexation of the IL with CO2, by using aprotic heterocyclic anions (AHA ILs) that eliminate the pervasive hydrogen bonding and salt bridge formation that is the origin of the viscosity increase.  Third, we will describe the discovery of AHA ILs whose melting points when reacted with CO2 are more than 100 °C below the melting point of the unreacted material.  These materials allow one to dramatically reduce the energy required for CO2 release and regeneration of the absorption material because a significant amount of the energy needed for the regeneration comes from the heat of fusion as the material releases CO2 and turns from liquid to solid.  Currently, we are working to combat the mass transfer challenges associated with the high viscosity of ILs by encapsulating them in polymeric shells.

April 22, 2019
10:30 AM
Room 1011, EB1
Dr. Yi Tang
UCLA

For decades, fungi have been an important source of medically relevant natural products (NPs). Recent advances in DNA sequencing have revealed that the biosynthetic potential of fungal genomes is much deeper than previously realized. Difficulties in culturing and genetically engineering many fungi, combined with the fact that many NP biosynthetic gene clusters (BGCs) are not expressed under standard laboratory conditions has lead to much of this biosynthetic potential remaining untapped. Here we describe the realization of a pipeline based in S. cerevisiae encompassing bioinformatic tools for BGC curation, genetic parts for BGC refactoring, and improved DNA assembly for BGC building. With this pipeline, we have successfully detected novel NPs from several previously unstudied fungal BGCs, and have structurally characterized a subset of the BGC-associated compounds. We also developed activity-guided methods to discover natural products of new function, and validated the biological activity using higher-order model systems.  Our pipeline demonstrates how high-throughput synthetic biology tools can facilitate the rapid discovery of complex chemical scaffolds of potential pharmaceutical relevance and their production in model fungal hosts.

April 29, 2019
10:30 AM
Room 1025, EB2
Dr. James McAndrew
Air Liquide

I will introduce Air Liquide, its business and innovation activity, and will present some recent research related to the use of nitrogen to reduce water consumption in hydraulic fracturing..   Hydraulic fracturing has dramatically augmented oil and gas production in North America, leading to improved energy independence and increasing the use of natural gas for electricity generation, with a corresponding decrease in carbon emissions. However,  hydraulic fracturing also has negative aspects, including the consumption of  large amounts of water, often in water-stressed regions with limited infrastructure.  The resulting disposal of produced water has led to earthquakes and contamination of surface water.  Nitrogen foams allow substantial reduction in water consumption and disposal. Widespread commercial implementation of nitrogen foams requires that production benefits also be demonstrated.  Therefore, we have constructed a laboratory apparatus to evaluate the benefits of nitrogen foams in terms of proppant transport.  As will be discussed, proppant transport improvements contribute to productivity of fractured reservoirs. Proppant transport characteristics of nitrogen foams will be compared with those of water.  The use of lower concentrations of nitrogen, to create “energized fluids” (bubbly liquids) will also be discussed, with recent results for proppant transport performance of proppant with a hydrophobic coating.