A Design Thinking Approach to Teaching Sustainability

Understanding sustainability requires a system-wide perspective to guide the interpretation of problems and conceptualization of solutions. A lake sustainability Trust provided an opportunity for students to gain this perspective by examining societal, commercial, and environmental policy perspectives associated with a local endangered lake and surrounding wetlands. This was an ideal project to bring sustainability issues to life for students but was too complex for a single class to attack. This article describes a multicourse initiative that sought to heighten students’ awareness of sustainability issues using a design thinking problem-solving approach. Interviews conducted revealed concerns that educators may have in considering similar projects. The authors respond to these perceived obstacles with recommendations and a discussion of mitigation strategies. In addition to the Trust gaining direction, the design and implementation of this cross-course experiential learning initiative allowed Marketing Research and Product Design students to develop an appreciation for macrolevel sustainability issues, and environmental policy students to appreciate the value of marketing research in the development of land use plans.

Free diffusion of PI(4,5)P2 in the plasma membrane in the presence of high density effector protein complexes

The lipid phosphatidyl-D-myo-inositol-4,5-bisphosphate [PI(4,5)P2] is a master regulator of plasma membrane (PM) function. It engages effector proteins that regulate diverse traffic, transport, signaling and cytoskeletal processes that define PM structure and function. How a single class of lipid molecules independently regulate so many parallel processes remains an open question. We tested the hypothesis that spatially segregated pools of PI(4,5)P2 are associated with, and thus reserved for regulation of, different functional complexes in the PM. The mobility of PI(4,5)P2 in the membrane was measured using lipid biosensors by single particle tracking photoactivation localization microscopy (sptPALM). We found that PI(4,5)P2, and several other classes of inner PM lipids, diffuse rapidly at approximately 0.3 microns squared per second with largely Brownian motion, although they spend approximately a third of their time diffusing much more slowly. Surprisingly, areas of the PM occupied by PI(4,5)P2-dependent complexes, such endoplasmic-reticulum:PM contact sites, clathrin-coated structures, and several actin cytoskeletal elements including focal adhesions, did not cause a change in PI(4,5)P2 lateral mobility. Only the spectrin and septin cytoskeletons were observed to produce a slowing of PI(4,5)P2 diffusion. We conclude that even structures with high densities of PI(4,5)P2-engaging effector proteins, such as clathrin coated pits and focal adhesions, do not corral free PI(4,5)P2, questioning a role for spatially segregated PI(4,5)P2 pools in organizing and regulating parallel PM functions.

Living Cells, Lifeless Molecules

Cells are composed of molecules that are lifeless but special, because most of them occur in nature only in the context of life. They are essential to all the workings of life, and no one single class holds life’s secret: life is an emergent property of the collective of molecules, assembled into the elaborate structures called cells. Cells come in great profusion, but all are variations on just two patterns of organization: prokaryotes, small and relatively simple microbes, both Bacteria and Archaea; and eukaryotes (Eukarya), the larger and more complex cells that make up all animals, plants, and fungi. The molecules of life, for all their diversity, again fall mainly into just a handful of categories. The bulk of living matter consists of proteins, nucleic acids, carbohydrates and lipids. Biomolecules belong to chemistry, but their functions in the process of living place them in the realm of biology.

A Probabilistic Approach to Growth Networks

Single-class closed queueing networks, consisting of infinite-server and single-server queues with exponential service times and probabilistic routing, admit product-from solutions. Such solutions, although seemingly tractable, are difficult to characterize explicitly for practically relevant problems due to the exponential combinatorial complexity of its normalization constant (partition function). In “A Probabilistic Approach to Growth Networks,” Jelenković, Kondev, Mohapatra, and Momčilović develop a novel methodology, based on a probabilistic representation of product-form solutions and large-deviations concentration inequalities, which identifies distinct operating regimes and yields explicit expressions for the marginal distributions of queue lengths. From a methodological perspective, a fundamental feature of the proposed approach is that it provides exact results for order-one probabilities, even though the analysis involves large-deviations rate functions, which characterize only vanishing probabilities on a logarithmic scale.

Detection of acne by deep learning object detection

ImportanceState-of-the art performance is achieved with a deep learning object detection model for acne detection. There is little current research on object detection in dermatology and acne in particular. As such, this work is early in this field and achieves state of the art performance.ObjectiveTrain an object detection model on a publicly available data set of acne photos.Design, Setting, and ParticipantsA deep learning model is trained with cross validation on a data set of facial acne photos.Main Outcomes and MeasuresObject detection models for detecting acne for single-class (acne) and multi-class (four severity levels). We train and evaluate the models using standard metrics such as mean average precision (mAP). Then we manually evaluate the model predictions on the test set, and calculate accuracy in terms of precision, recall, F1, true and false positive and negative detections.ResultsWe achieve state-of-the art mean average precision [email protected] value of 37.97 for the single class acne detection task, and 26.50 for the 4-class acne detection task. Moreover, our manual evaluation shows that the single class detection model performs well on the validation set, achieving true positive 93.59 %, precision 96.45 % and recall 94.73 %.Conclusions and RelevanceWe are able to train a high-accuracy acne detection model using only a small publicly available data set of facial acne. Transfer learning on the pre-trained deep learning model yields good accuracy and high degree of transferability to patient submitted photographs. We also note that the training of standard architecture object detection models has given significantly better accuracy than more intricate and bespoke neural network architectures in the existing research literature.Key PointsQuestionCan deep learning-based acne detection models trained on a small data set of publicly available photos of patients with acne achieve high prediction accuracy?FindingsWe find that it is possible to train a reasonably good object detection model on a small, annotated data set of acne photos using standard deep learning architectures.MeaningDeep learning-based object detection models for acne detection can be a useful decision support tools for dermatologists treating acne patients in a digital clinical practice. It can prove a particularly useful tool for monitoring the time evolution of the acne disease state over prolonged time during follow-ups, as the model predictions give a quantifiable and comparable output for photographs over time. This is particularly helpful in teledermatological consultations, as a prediction model can be integrated in the patient-doctor remote communication.

Development of PRPK Directed Phthalimides

Immunomodulatory drugs (IMiDs) thalidomide, lenalidomide, and pomalidomide (Pom) bind to cereblon (CRBN) and trigger proteasomal degradation of neo-substrates IKZF1/3 leading to multiple myeloma (MM) cell apoptosis. Pomalidomide (Pom) also binds albeit weakly to p53-related protein kinase (PRPK, aka TP53RK), an understudied kinase reported to be associated with poor prognosis in MM patients. Here, we developed a series of IMiDs based on Pom and conducted a structure-activity relationship (SAR) study to identify a potent and selective PRPK binder. Structural analysis showed that IMiDs bind PRPK in a fundamentally different way from CRBN, and suggested specific derivatization to improve affinity. We employed a structure-guided strategy to develop compound TXM-02-118, which exhibited nanomolar affinityfor PRPK in binding assays, and showed high selectivity for PRPK over CRBN. Overall, the work represents an initial effort to develop tool compounds for studying PRPK. Moreover, it illustrates how a single class of molecules can use different recognition elements to bind diverse targets using fundamentally different binding poses. This has broad implications for chemical probe and lead compound selectivity profiling, and argues for more wide-spread use of global proteomics or similar methodologies.

Electrochemical Organic Synthesis of Electron-Rich Biaryl Scaffolds: An Update

Biaryl scaffolds are widely spread in biologically important natural products, in numerous therapeutic agents, but they are also considered a privileged class of ligands and (organo)catalysts; therefore, the development of efficient alternative methodologies to prepare such compounds is always attracting much attention. The present review discusses the organic electrosynthesis of biaryls starting from phenols, anilines, naphthols, and naphthylamines. The most significant examples of the works reported in the last decade are presented and classified according to the single class of molecules: after the introduction, the first three sections relate to the reactions of phenols, naphthols, and anilines, respectively; the other two sections refer to cross-coupling and miscellaneous reactions.

Distinctive Kinesin-14 Motors Associate with Midzone Microtubules to Construct Mitotic Spindles with Two Convergent Poles in Arabidopsis

Microtubule (MT) motors in the Kinesin-14 subfamily proliferated in photosynthetic organisms and they often incorporated sequences bearing novel structural features. To gain insights into the functions of diversified Kinesin-14 motors from an evolutionary perspective, we performed phylogenetic analyses across different eukaryotic kingdoms. Compared to fungi that have a single class of Kinesin-14, the early divergent protist Giardia possesses two classes and the motile green alga Chlamydomonas produces four classes (Kinesin-14A to Kinesin-14D). The fifth class Kinesin-14E first appeared among immotile green algae and the sixth Kinesin-14F emerged in mosses, concomitantly with the display of 3D growth. The conservation of Kinesin-14D from green algae prompted us to investigate its function in Arabidopsis in which three such motors functioned in cell cycle-dependent manners. They localized on selective spindle MTs and/or sometimes kinetochore-like structures, and later all became conspicuous on MT bundles in the spindle midzone following sister chromatid segregation. Genetic dissection of Kinesin-14D1 showed that its loss led to hypersensitivity to low doses of the MT-depolymerizing herbicide oryzalin. Kinesin-14D1 association with the midzone MTs in both prophase and mitotic spindles. The oryzalin treatment left behind discrete kinetochore fibers attached to randomly positioned chromosomes in the mitotic kinesin-14d1 cells but prevented the pole convergence of bipolar mitotic spindles. This function of Kinesin-14D1 in the spindle midzone is likely dependent on an MT-binding domain at the C-terminus to the catalytic motor domain. Therefore, our results revealed a novel Kinesin-14D-dependent mechanism that regulates the formation of bipolar spindle apparatus with converged acentrosomal poles.

OwnKit: Ownership Inference for Java

<p>Object ownership allows us to statically control run-time aliasing in order to provide a strong notion of object encapsulation. Unfortunately in order to use ownership, code must first be annotated with extra type information. This imposes a heavy burden on the programmer, and has contributed to the slow adoption of ownership. Ownership inference is the process of reconstructing ownership type information based on the existing ownership patterns in code. This thesis presents OwnKit—an automatic ownership inference tool for Java. OwnKit conducts inference in a modular way: by only considering a single class at the time. The modularity makes our algorithm highly scalable in both time and memory usage.</p>

OwnKit: Ownership Inference for Java

<p>Object ownership allows us to statically control run-time aliasing in order to provide a strong notion of object encapsulation. Unfortunately in order to use ownership, code must first be annotated with extra type information. This imposes a heavy burden on the programmer, and has contributed to the slow adoption of ownership. Ownership inference is the process of reconstructing ownership type information based on the existing ownership patterns in code. This thesis presents OwnKit—an automatic ownership inference tool for Java. OwnKit conducts inference in a modular way: by only considering a single class at the time. The modularity makes our algorithm highly scalable in both time and memory usage.</p>