Publications

We present a fast direct solver for the simulation of electromagnetic scattering from an arbitrarily-shaped, large, empty cavity embedded in an infinite perfectly conducting half space. The governing Maxwell equations are reformulated as a well-conditioned second kind integral equation and the resulting linear system is solved in nearly linear time using a hierarchical matrix factorization technique. We illustrate the performance of the scheme with several numerical examples for complex cavity shapes over a wide range of frequencies.

Olshausen and Field (OF) proposed that neural computations in the primary visual cortex (V1) can be partially modelled by sparse dictionary learning. By minimizing the regularized representation error they derived an online algorithm, which learns Gabor-filter receptive fields from a natural image ensemble in agreement with physiological experiments. Whereas the OF algorithm can be mapped onto the dynamics and synaptic plasticity in a single-layer neural network, the derived learning rule is nonlocal - the synaptic weight update depends on the activity of neurons other than just pre- and postsynaptic ones - and hence biologically implausible. Here, to overcome this problem, we derive sparse dictionary learning from a novel cost-function - a regularized error of the symmetric factorization of the input's similarity matrix. Our algorithm maps onto a neural network of the same architecture as OF but using only biologically plausible local learning rules. When trained on natural images our network learns Gabor-filter receptive fields and reproduces the correlation among synaptic weights hard-wired in the OF network. Therefore, online symmetric matrix factorization may serve as an algorithmic theory of neural computation.

Social media use among elites offers a useful avenue for analyzing regime response to protest, especially in countries with some degree of freedom of speech. We examine the frequency and content of Twitter usage among Venezuelan elites in the context of the 2014 protests. This analysis demonstrates that the regime sent more signals during protests but that the content of these messages addressed more topics than it did for opposition elites, especially following acts of regime suppression of opposition-sponsored protests. This observation supports theoretical predictions that noisy public information can make coordination more difficult. Regime elites produced noisy Twitter as part of an explicit strategy to prevent citizens from coordinating on revolution. This analysis of social media use by both pro- and anti-regime elites contributes to the debate over whether social media prevents or promotes regime change.

Peptomers are oligomeric molecules composed of both α-amino acids and N-substituted glycine monomers, thus creating a hybrid of peptide and peptoid units. Peptomers have been used in several applications such as antimicrobials, protease inhibitors, and antibody mimics. Despite the considerable promise of peptomers as chemically diverse molecular scaffolds, we know little about their conformational tendencies. This lack of knowledge limits the ability to implement computational approaches for peptomer design. Here we computationally evaluate the local structural propensities of the peptide–peptoid linkage. We find some general similarities between the peptide residue conformational preferences and the Ramachandran distribution of residues that precede proline in folded protein structures. However, there are notable differences. For example, several β-turn motifs are disallowed when the i+2 residue is also a peptoid monomer. Significantly, the lowest energy geometry, when dispersion forces are accounted for, corresponds to a “cis-Pro touch-turn” conformation, an unusual turn motif that has been observed at protein catalytic centers and binding sites. The peptomer touch-turn thus represents a useful design element for the construction of folded oligomers capable of molecular recognition and as modules in the assembly of structurally complex peptoid–protein hybrid macromolecules.

Are legislators responsive to their constituents in their public communication? To what extent are they able to shape the agenda that the mass public cares about, as expressed by the issues they discuss? We address this twofold question with an analysis of all tweets sent by Members of the U.S. Congress and a random sample of their followers from January  to March . Using a Latent Dirichlet Allocation model, we extract topics that represent the diversity of issues that legislators and ordinary citizens discuss on this social networking site. en, we exploit variation in the distribution of topics over time to test whether Members of Congress lead or follow their constituents in their selection of issues to discuss, employing a Granger-causality framework. We find that legislators are responsive in their public statements to their constituents, but also that they have limited influence on their followers’ public agenda. To further understand the mechanisms that explain political responsiveness, we also examine whether Members of Congress are more responsive to specific constituents groups, showing that they are more influenced by co-partisans, politically interested citizens, and social media users located within their constituency.

With the notable exception of B-cell malignancies, the efficacy of chimeric antigen receptor (CAR) T cells has been limited, and CAR T cells have not been shown to expand and persist in patients with nonlymphoid tumors. Here we demonstrate that redirection of primary human T cells with a CAR containing the inducible costimulator (ICOS) intracellular domain generates tumor-specific IL-17-producing effector cells that show enhanced persistence. Compared with CARs containing the CD3ζ chain alone, or in tandem with the CD28 or the 4-1BB intracellular domains, ICOS signaling increased IL-17A, IL-17F, and IL-22 following antigen recognition. In addition, T cells redirected with an ICOS-based CAR maintained a core molecular signature characteristic of TH17 cells and expressed higher levels of RORC, CD161, IL1R-1, and NCS1. Of note, ICOS signaling also induced the expression of IFN-γ and T-bet, consistent with a TH17/TH1 bipolarization. When transferred into mice with established tumors, TH17 cells that were redirected with ICOS-based CARs mediated efficient antitumor responses and showed enhanced persistence compared with CD28- or 4-1BB-based CAR T cells. Thus, redirection of TH17 cells with a CAR encoding the ICOS intracellular domain is a promising approach to augment the function and persistence of CAR T cells in hematologic malignancies.

Cells respond to environmental stimuli with expression changes at both the mRNA and protein level, and a plethora of known and unknown regulators affect synthesis and degradation rates of the resulting proteins. To investigate the major principles of gene expression regulation in dynamic systems, we estimated protein synthesis and degradation rates from parallel time series data of mRNA and protein expression and tested the degree to which expression changes can be modeled by a simple linear differential equation. Examining three published datasets for yeast responding to diamide, rapamycin, and sodium chloride treatment, we find that almost one-third of genes can be well-modeled, and the estimated rates assume realistic values. Prediction quality is linked to low measurement noise and the shape of the expression profile. Synthesis and degradation rates do not correlate within one treatment, consistent with their independent regulation. When performing robustness analyses of the rate estimates, we observed that most genes adhere to one of two major modes of regulation, which we term synthesis- and degradation-independent regulation. These two modes, in which only one of the rates has to be tightly set, while the other one can assume various values, offer an efficient way for the cell to respond to stimuli and re-establish proteostasis. We experimentally validate degradation-independent regulation under oxidative stress for the heatshock protein Ssa4.

Recently we demonstrated that incorporating p-fluorophenylalanine (pFF) into phosphotriesterase dramatically improved folding, thereby leading to enhanced stability and function at elevated temperatures. To further improve the stability of the fluorinated enzyme, Rosetta was used to identify multiple potential stabilizing mutations. One such variant, pFF-F104A, exhibited enhanced activity at elevated temperature and maintained activity over many days in solution at room temperature.

Interleukin (IL)-22–producing group 3 innate lymphoid cells (ILC3) promote mucosal healing and maintain barrier integrity, but how microbial signals are integrated to regulate mucosal protection offered by these cells remains unclear. Here, we show that in vivo depletion of CX3CR1+ mononuclear phagocytes (MNPs) resulted in more severe colitis and death after infection with Citrobacter rodentium. This phenotype was rescued by exogenous IL-22, which was endogenously produced by ILC3 in close spatial proximity to CX3CR1+ MNPs that were dependent on MyD88 signaling. CX3CR1+ MNPs from both mouse and human tissue produced more IL-23 and IL-1β than conventional CD103+ dendritic cells (cDCs) and were more efficient than cDCs in supporting IL-22 production in ILC3 in vitro and in vivo. Further, colonic ILC3 from patients with mild to moderate ulcerative colitis or Crohn’s disease had increased IL-22 production. IBD-associated SNP gene set analysis revealed enrichment for genes selectively expressed in human intestinal MNPs. The product of one of these, TL1A, potently enhanced IL-23– and IL-1β-induced production of IL-22 and GM-CSF by ILC3. Collectively, these results reveal a critical role for CX3CR1+ mononuclear phagocytes in integrating microbial signals to regulate colonic ILC3 function in IBD.