Rat corneal endothelial cell migration during wound repair on the basement membrane depends more on the PI-3K pathway than the cdc-42 pathway or actin stress fibers.

Following a central transcorneal circular freeze injury, organ-cultured rat corneal endothelial cells surrounding the wound reorganize peripheral actin bands into stress fibers and migrate individually into the wound. To ascertain the significance of this rearrangement relative to morphological changes accompanying migration and wound repair, some tissues were incubated overnight in 4 µM TRITC-conjugated phalloidin to stabilize actin and prevent its reorganization. After a freeze injury to the endothelium tissues were histologically observed at 24 h post-wounding and demonstrated that despite a lack of actin organization, cells responding to the injury appeared morphologically similar to their control counterparts. Tissues cultivated in the presence of either cytochalasin B (CB), soybean agglutinin (SBA), or fluorouracil (FU) and also exhibited actin cytoskeletal disruption. Under these conditions, migration continued despite the absence of detectable stress fibers. For SBA-, CB-, and FU-treated tissues, wound repair did not significantly differ from control preparations although FU-treated tissues showed a slower repair. Electron micrographs confirmed an absence of stress fibers in migrating cells treated with any of these agents. Tissues were also treated with ML 141 and EY294002 to inhibit the cdc-42 and PI-3K pathways, respectively. While cell movement still occurred in the presence of ML 141, migration into the wound was greatly restricted in the presence of EY294002. These results indicate that rat corneal endothelial cell movement in situ does not require actin reorganization into stress fibers, but the functioning of the PI-3K pathway is indispensable for their migration along the basement membrane during wound repair.

Maternal obesity is not associated with increased crown-rump length in first-trimester ultrasounds of singleton gestations.

OBJECTIVE: To evaluate whether or not obesity affects fetal growth in the first trimester of pregnancy. STUDY DESIGN: A retrospective cohort study of obese versus non-obese women in our ultrasound database was performed to compare crown-rump length (CRL), a surrogate of fetal growth, at the first-trimester genetic screening. RESULTS: A total of 50 obese and 50 non-obese women were included. CRL for both groups was performed at an average of 12wk5d ± 3 d. A linear regression analysis demonstrated that there was no difference between the cohorts in respect to CRL in the first trimester (p = .482). However, the estimated fetal weight at second-trimester anatomy ultrasound and the neonatal birth weight were increased in obese women (p < .001 for both analyses). CONCLUSION: Maternal obesity does not significantly alter the fetal CRL. However, maternal obesity appears to be associated with increased fetal growth as early as the second trimester.

Androgens in maternal vascular and placental function: implications for preeclampsia pathogenesis

Adequate maternal vascular adaptations and blood supply to the uterus and placenta are crucial for optimal oxygen and nutrient transport to growing fetuses of eutherian mammals, including humans. Multiple factors contribute to hemodynamics and structuring of placental vasculature essential for term pregnancy with minimal complications. In women, failure to achieve or sustain favorable pregnancy progression is, not surprisingly, associated with high incidence of antenatal complications, including preeclampsia, a hypertensive disorder of pregnancy. While the pathogenesis of preeclampsia in women remains unknown, a role for androgens is emerging. The relationship between androgens and maternal cardiovascular and placental function deserves particular consideration because testosterone levels in the circulation of preeclamptic women are elevated approximately two- to three-fold and are positively correlated with vascular dysfunction. Preeclampsia is also associated with elevated placental androgen receptor (AR) gene expression. Studies in animal models mimicking the pattern and level of increase of adult female testosterone levels to those found in preeclamptic pregnancies, replicate key features of preeclampsia, including gestational hypertension, endothelial dysfunction, exaggerated vasoconstriction to angiotensin II, reduced spiral artery remodeling, placental hypoxia, decreased nutrient transport and fetal growth restriction. Taken together, these data strongly implicate AR-mediated testosterone action as an important pathway contributing to clinical manifestation of preeclampsia. This review critically addresses this hypothesis, taking into consideration both clinical and preclinical data.

A Density Functional Tight Binding Layer for Deep Learning of Chemical Hamiltonians.

Current neural networks for predictions of molecular properties use quantum chemistry only as a source of training data. This paper explores models that use quantum chemistry as an integral part of the prediction process. This is done by implementing self-consistent-charge Density-Functional-Tight-Binding (DFTB) theory as a layer for use in deep learning models. The DFTB layer takes, as input, Hamiltonian matrix elements generated from earlier layers and produces, as output, electronic properties from self-consistent field solutions of the corresponding DFTB Hamiltonian. Backpropagation enables efficient training of the model to target electronic properties. Two types of input to the DFTB layer are explored, splines and feed-forward neural networks. Because overfitting can cause models trained on smaller molecules to perform poorly on larger molecules, regularizations are applied that penalize nonmonotonic behavior and deviation of the Hamiltonian matrix elements from those of the published DFTB model used to initialize the model. The approach is evaluated on 15 700 hydrocarbons by comparing the root-mean-square error in energy and dipole moment, on test molecules with eight heavy atoms, to the error from the initial DFTB model. When trained on molecules with up to seven heavy atoms, the spline model reduces the test error in energy by 60% and in dipole moments by 42%. The neural network model performs somewhat better, with error reductions of 67% and 59%, respectively. Training on molecules with up to four heavy atoms reduces performance, with both the spline and neural net models reducing the test error in energy by about 53% and in dipole by about 25%.

Constant size descriptors for accurate machine learning models of molecular properties.

Two different classes of molecular representations for use in machine learning of thermodynamic and electronic properties are studied. The representations are evaluated by monitoring the performance of linear and kernel ridge regression models on well-studied data sets of small organic molecules. One class of representations studied here counts the occurrence of bonding patterns in the molecule. These require only the connectivity of atoms in the molecule as may be obtained from a line diagram or a SMILES string. The second class utilizes the three-dimensional structure of the molecule. These include the Coulomb matrix and Bag of Bonds, which list the inter-atomic distances present in the molecule, and Encoded Bonds, which encode such lists into a feature vector whose length is independent of molecular size. Encoded Bonds' features introduced here have the advantage of leading to models that may be trained on smaller molecules and then used successfully on larger molecules. A wide range of feature sets are constructed by selecting, at each rank, either a graph or geometry-based feature. Here, rank refers to the number of atoms involved in the feature, e.g., atom counts are rank 1, while Encoded Bonds are rank 2. For atomization energies in the QM7 data set, the best graph-based feature set gives a mean absolute error of 3.4 kcal/mol. Inclusion of 3D geometry substantially enhances the performance, with Encoded Bonds giving 2.4 kcal/mol, when used alone, and 1.19 kcal/mol, when combined with graph features.

Foetal premature atrial contractions during the second and third trimester are not associated with foetal breathing.

Foetal premature atrial contractions (PACs) are the most commonly encountered and also the most benign foetal arrhythmia. A retrospective cohort study was conducted with the objective to assess whether the presence of foetal breathing was associated with the presence of foetal PACs. A further objective was to evaluate whether this association would affect neonatal outcomes at a high volume referral centre. The diagnosis of PACs was based on the observation of a premature atrial contraction followed by a ventricular contraction on ultrasound myocardial M-mode. Trained ultrasonographers documented in the ultrasound report whether or not foetal breathing was present with PACs. 91 exams were identified, which included 75 individual pregnancies. Six women were identified who had foetal PACs associated with foetal breathing on ultrasound evaluation. Foetuses with PACs did not differ between the associated breathing and no-associated breathing groups with respect to maternal age, parity, mode of delivery, gestational age at delivery or birthweight. This study reaffirms that isolated PACs are a benign finding. Furthermore, it adds to the pool of literature on foetal PACs in that it is not associated with abnormal pregnancy outcomes regardless of the presence or absence of foetal breathing. Impact statement • What is already known on this subjectSince foetal breathing can effect Doppler ultrasound assessment of the foetal cardiovascular system, it is reasonable to consider that it may impact conditions such as foetal arrhythmias. • What the results of this study addFoetal breathing does not impact on the presence of premature atrial contractions. • What the implications are of these findings for clinical practice and/or further researchFoetal breathing is not associated with the finding of foetal premature atrial contractions.

Supervised Learning for Dynamical System Learning.

Recently there has been substantial interest in spectral methods for learning dynamical systems. These methods are popular since they often offer a good tradeoff between computational and statistical efficiency. Unfortunately, they can be difficult to use and extend in practice: e.g., they can make it difficult to incorporate prior information such as sparsity or structure. To address this problem, we present a new view of dynamical system learning: we show how to learn dynamical systems by solving a sequence of ordinary supervised learning problems, thereby allowing users to incorporate prior knowledge via standard techniques such as L1 regularization. Many existing spectral methods are special cases of this new framework, using linear regression as the supervised learner. We demonstrate the effectiveness of our framework by showing examples where nonlinear regression or lasso let us learn better state representations than plain linear regression does; the correctness of these instances follows directly from our general analysis.

Clinician confidence about conversations at the end of life is strengthened using the four habits approach.

Can attending a communication skills workshop focused on advance care planning, shifting focus to palliative care, personal grief, managing anger, and culture and communication result in changes in attitudes and knowledge? One hundred and three clinicians completed surveys prior to and following the workshop resulting in significant changes in knowledge, attitudes and intent to change that persisted for at least 3 months. As most clinicians are not routinely exposed to learning communication skills for end-of-life conversations, opportunities to practice these skills in a safe supportive environment should be made available in medical schools, residency programs, and in the practice community.

Automated image analysis of protein localization in budding yeast.

MOTIVATION: The yeast Saccharomyces cerevisiae is the first eukaryotic organism to have its genome completely sequenced. Since then, several large-scale analyses of the yeast genome have provided extensive functional annotations of individual genes and proteins. One fundamental property of a protein is its subcellular localization, which provides critical information about how this protein works in a cell. An important project therefore was the creation of the yeast GFP fusion localization database by the University of California, San Francisco, USA (UCSF). This database provides localization data for 75% of the proteins believed to be encoded by the yeast genome. These proteins were classified into 22 distinct subcellular location categories by visual examination. Based on our past success at building automated systems to classify subcellular location patterns in mammalian cells, we sought to create a similar system for yeast. RESULTS: We developed computational methods to automatically analyze the images created by the UCSF yeast GFP fusion localization project. The system was trained to recognize the same location categories that were used in that study. We applied the system to 2640 images, and the system gave the same label as the previous assignments to 2139 images (81%). When only the highest confidence assignments were considered, 94.7% agreement was observed. Visual examination of the proteins for which the two approaches disagree suggests that at least some of the automated assignments may be more accurate. The automated method provides an objective, quantitative and repeatable assignment of protein locations that can be applied to new collections of yeast images (e.g. for different strains or the same strain under different conditions). It is also important to note that this performance could be achieved without requiring colocalization with any marker proteins. AVAILABILITY: The original images analyzed in this article are available at http://yeastgfp.ucsf.edu, and source code and results are available at http://murphylab.web.cmu.edu/software.

Not the same everywhere. Patient-centered learning environments at nine medical schools.

BACKGROUND: Learning environments overtly or implicitly address patient-centered values and have been the focus of research for more than 40 years, often in studies about the "hidden curriculum." However, many of these studies occurred at single medical schools and used time-intensive ethnographic methods. This field of inquiry lacks survey methods and information about how learning environments differ across medical schools. OBJECTIVE: To examine patient-centered characteristics of learning environments at 9 U.S. medical schools. DESIGN: Cross-sectional internet-based survey. PARTICIPANTS: Eight-hundred and twenty-three third- and fourth-year medical students in the classes of 2002 and 2003. MEASUREMENTS: We measured the patient-centeredness of learning environments with the Communication, Curriculum, and Culture (C3) Instrument, a 29-item validated measure that characterizes the degree to which a medical school's environment fosters patient-centered care. The C3 Instrument contains 3 content areas (role modeling, students' experiences, and support for students' patient-centered behaviors), and is designed to measure these areas independent of respondents' attitudes about patient-centered care. We also collected demographic and attitudinal information from respondents. RESULTS: The variability of C3 scores across schools in each of the 3 content areas of the instrument was striking and statistically significant (P values ranged from .001 to .004). In addition, the patterns of scores on the 3 content areas differed from school to school. CONCLUSIONS: The 9 schools demonstrated unique and different learning environments both in terms of magnitude and patterns of characteristics. Further multiinstitutional study of hidden curricula is needed to further establish the degree of variability that exists, and to assist educators in making informed choices about how to intervene at their own schools.

Assessing competence in communication and interpersonal skills: the Kalamazoo II report.

Accreditation of residency programs and certification of physicians requires assessment of competence in communication and interpersonal skills. Residency and continuing medical education program directors seek ways to teach and evaluate these competencies. This report summarizes the methods and tools used by educators, evaluators, and researchers in the field of physician-patient communication as determined by the participants in the "Kalamazoo II" conference held in April 2002. Communication and interpersonal skills form an integrated competence with two distinct parts. Communication skills are the performance of specific tasks and behaviors such as obtaining a medical history, explaining a diagnosis and prognosis, giving therapeutic instructions, and counseling. Interpersonal skills are inherently relational and process oriented; they are the effect communication has on another person such as relieving anxiety or establishing a trusting relationship. This report reviews three methods for assessment of communication and interpersonal skills: (1) checklists of observed behaviors during interactions with real or simulated patients; (2) surveys of patients' experience in clinical interactions; and (3) examinations using oral, essay, or multiple-choice response questions. These methods are incorporated into educational programs to assess learning needs, create learning opportunities, or guide feedback for learning. The same assessment tools, when administered in a standardized way, rated by an evaluator other than the teacher, and using a predetermined passing score, become a summative evaluation. The report summarizes the experience of using these methods in a variety of educational and evaluation programs and presents an extensive bibliography of literature on the topic. Professional conversation between patients and doctors shapes diagnosis, initiates therapy, and establishes a caring relationship. The degree to which these activities are successful depends, in large part, on the communication and interpersonal skills of the physician. This report focuses on how the physician's competence in professional conversation with patients might be measured. Valid, reliable, and practical measures can guide professional formation, determine readiness for independent practice, and deepen understanding of the communication itself.

Care not cure: dialogues at the transition.

Physicians and patients find it hard to communicate when treatment fails to cure or control cancer. Communication barriers include fear of "giving up," losing the medical team, and discussing death. The quality of physician-patient communication affects important outcomes including patient distress, coping, and quality of life, and physician burnout. Communication skills that can be taught, learned, and maintained for physicians at all levels of training, and effective educational programs have been described. Research on communication skills training should focus on the best method of delivery, the "dose-response" effect, and how to measure success of training in complex health care environments.

Production of beta-glucosidase using immobilised Piromyces sp. KSX1 and Orpinomyces sp. 478P1 in repeat-batch culture.

Two anaerobic fungi, one a monocentric strain ( Piromyces sp. KSX1) and the other a polycentric strain ( Orpinomyces sp. 478P1), were immobilised in calcium alginate beads and cultured in sequential batches where spent medium (containing 0.25% cellobiose) was repeatedly drained and replaced. beta-Glucosidase production with KSX1 was maintained for 45 days over six repeated batch cultures yielding a maximum level of 107 mIU/ml. For 478P1, beta-glucosidase production was maintained for 30 days over four repeated batches yielding a maximum level of 34 mIU/ml. Although repeat-batch cultures of KSX1 produced more beta-glucosidase than strain 478P1, the maximum specific beta-glucosidase produced from these immobilised cultures was similar. The immobilised polycentric strain proved to be operationally superior to strain KSX1, as strain 478P1 did not produce any growth in the culture liquor.