Measuring the prediction difficulty of individual cases in a dataset using machine learning.

Different levels of prediction difficulty are one of the key factors that researchers encounter when applying machine learning to data. Although previous studies have introduced various metrics for assessing the prediction difficulty of individual cases, these metrics require specific dataset preconditions. In this paper, we propose three novel metrics for measuring the prediction difficulty of individual cases using fully-connected feedforward neural networks. The first metric is based on the complexity of the neural network needed to make a correct prediction. The second metric employs a pair of neural networks: one makes a prediction for a given case, and the other predicts whether the prediction made by the first model is likely to be correct. The third metric assesses the variability of the neural network's predictions. We investigated these metrics using a variety of datasets, visualized their values, and compared them to fifteen existing metrics from the literature. The results demonstrate that the proposed case difficulty metrics were better able to differentiate various levels of difficulty than most of the existing metrics and show constant effectiveness across diverse datasets. We expect our metrics will provide researchers with a new perspective on understanding their datasets and applying machine learning in various fields.

Solution-based Supramolecular Hierarchical Assembly of Frenkel Excitonic Nanotubes Driven by Gold Nanoparticle Formation and Temperature.

Translating nature's successful design principle of solution-based supramolecular self-assembling to broad applications─ranging from renewable energy and information technology to nanomedicine─requires a fundamental understanding of supramolecular hierarchical assembly. Though the forces behind self-assembly (e.g., hydrophobicity) are known, the specific mechanism by which monomers form the hierarchical assembly still remains an open question. A crucial step toward formulating a complete mechanism is understanding not only how the monomer's specific molecular structure but also how manifold environmental conditions impact the self-assembling process. Here, we elucidate the complex correlation between the environmental self-assembling conditions and the resulting structural properties by utilizing a well-characterized model system: well-defined supramolecular Frenkel excitonic nanotubes (NTs), self-assembled from cyanine dye molecules in aqueous solution, which further self-assemble into bundled nanotubes (b-NTs). The NTs and b-NTs inhabit distinct spectroscopic signatures, which allows the use of steady-state absorption spectroscopy to monitor the transition from NTs to b-NTs directly. Specifically, we investigate the impact of temperature (ranging from 23 °C, 55 °C, 70 °C, 85 °C, up to 100 °C) during in situ formation of gold nanoparticles to determine their role in the formation of b-NTs. The considered time regime for the self-assembling process ranges from 1 min to 8 days. With our work, we contribute to a basic understanding of how environmental conditions impact solution-based hierarchical supramolecular self-assembly in both the thermodynamic and the kinetic regime.

Synthesis, Structure, and Photophysical Properties of Platinum Compounds with Thiophene-Derived Cyclohexyl Diimine Ligands.

Platinum(II) and platinum(IV) compounds were prepared by the stereoselective and regioselective reactions of thiophene-derived cyclohexyl diimine C^N^N-ligands with [Pt2Me4(µ-SMe2)2]. Newly synthesized ligands were characterized by NMR spectroscopy and elemental analysis, and Pt(II)/Pt(IV) compounds were characterized by NMR spectroscopy, elemental analysis, high-resolution mass spectrometry, and single-crystal X-ray diffraction. UV-vis absorbance and photoluminescence measurements were performed on newly synthesized complexes, as well as structurally related Pt(II)/Pt(IV) compounds with benzene-derived cyclohexyl diimine ligands, in dichloromethane solution, as solids, and as 5% by weight PMMA-doped films. DFT and TD-DFT calculations were performed, and the results were compared with the observed spectroscopic properties of the newly synthesized complexes. X-ray total scattering measurements and real space pair distribution function analysis were performed on the synthesized complexes to examine the local- and intermediate-range atomic structures of the emissive solid states.

In situ synchrotron X-ray total scattering measurements and analysis of colloidal CsPbX3 nanocrystals during flow synthesis.

In situ X-ray scattering measurements of CsPbX3 (X = Cl, Br, I) nanocrystal formation and halide exchange at NSLS-II beamlines were performed in an automated flow reactor. Total scattering measurements were performed at the 28-ID-2 (XPD) beamline and small-angle X-ray scattering at the 16-ID (LiX) beamline. Nanocrystal structural parameters of interest, including size, size distribution and atomic structure, were extracted from modeling the total scattering data. The results highlight the potential of these beamlines and the measurement protocols described in this study for studying dynamic processes of colloidal nanocrystal synthesis in solution with timescales on the order of seconds.

COF-300 synthesis and colloidal stabilization with substituted benzoic acids.

Colloidal covalent organic framework (COF) synthesis enables morphological control of crystallite size and shape. Despite numerous examples of 2D COF colloids with various linkage chemistries, 3D imine-linked COF colloids are more challenging synthetic targets. Here we report a rapid (15 min-5 day) synthesis of hydrated COF-300 colloids ranging in length (251 nm-4.6 µm) with high crystallinity and moderate surface areas (150 m2 g-1). These materials are characterized by pair distribution function analysis, which is consistent with the known average structure for this material alongside different degrees of atomic disorder at different length scales. Additionally, we investigate a series of para-substituted benzoic acid catalysts, finding that 4-cyano and 4-fluoro substituted benzoic acids produce the largest COF-300 crystallites with lengths of 1-2 µm. In situ dynamic light scattering experiments are used to assess time to nucleation in conjunction with 1H NMR model compound studies to probe the impact of catalyst acidity on the imine condensation equilibrium. We observe cationically stabilized colloids with a zeta potential of up to +14.35 mV in benzonitrile as a result of the carboxylic acid catalyst protonating surface amine groups. We leverage these surface chemistry insights to synthesize small COF-300 colloids using sterically hindered diortho-substituted carboxylic acid catalysts. This fundamental study of COF-300 colloid synthesis and surface chemistry will provide new insights into the role of acid catalysts both as imine condensation catalysts and as colloid stabilizing agents.

Teenage dating abuse, rape, and violence: The dermatologist's role.

Teenage dating abuse, rape, and violence are considered major public health problems that affect the lives of millions of teenagers in the United States. Dermatologists have traditionally become involved in these cases when confronted with patients who have unexplained bruising or other skin injuries and/or sexually transmitted diseases that raise the possibility that they could be victims of sexual abuse and violence. This contribution explores the role of the dermatologist in the diagnosis and management of teen dating abuse. We suggest some screening questions that might help to broach these serious issues with teen patients when the suspicion of dating abuse arises. We also provide a list of resources and hotlines that offer advice on how best to handle teen dating abuse. Some legal issues concerning the physician's role in managing teen dating abuse, rape, and violence are also discussed.

Dementia risk prediction in individuals with mild cognitive impairment: a comparison of Cox regression and machine learning models.

BACKGROUND: Cox proportional hazards regression models and machine learning models are widely used for predicting the risk of dementia. Existing comparisons of these models have mostly been based on empirical datasets and have yielded mixed results. This study examines the accuracy of various machine learning and of the Cox regression models for predicting time-to-event outcomes using Monte Carlo simulation in people with mild cognitive impairment (MCI). METHODS: The predictive accuracy of nine time-to-event regression and machine learning models were investigated. These models include Cox regression, penalized Cox regression (with Ridge, LASSO, and elastic net penalties), survival trees, random survival forests, survival support vector machines, artificial neural networks, and extreme gradient boosting. Simulation data were generated using study design and data characteristics of a clinical registry and a large community-based registry of patients with MCI. The predictive performance of these models was evaluated based on three-fold cross-validation via Harrell's concordance index (c-index), integrated calibration index (ICI), and integrated brier score (IBS). RESULTS: Cox regression and machine learning model had comparable predictive accuracy across three different performance metrics and data-analytic conditions. The estimated c-index values for Cox regression, random survival forests, and extreme gradient boosting were 0.70, 0.69 and 0.70, respectively, when the data were generated from a Cox regression model in a large sample-size conditions. In contrast, the estimated c-index values for these models were 0.64, 0.64, and 0.65 when the data were generated from a random survival forest in a large sample size conditions. Both Cox regression and random survival forest had the lowest ICI values (0.12 for a large sample size and 0.18 for a small sample size) among all the investigated models regardless of sample size and data generating model. CONCLUSION: Cox regression models have comparable, and sometimes better predictive performance, than more complex machine learning models. We recommend that the choice among these models should be guided by important considerations for research hypotheses, model interpretability, and type of data.

Persistent nucleation and size dependent attachment kinetics produce monodisperse PbS nanocrystals.

Modern syntheses of colloidal nanocrystals yield extraordinarily narrow size distributions that are believed to result from a rapid "burst of nucleation" (La Mer, JACS, 1950, 72(11), 4847-4854) followed by diffusion limited growth and size distribution focusing (Reiss, J. Chem. Phys., 1951, 19, 482). Using a combination of in situ X-ray scattering, optical absorption, and 13C nuclear magnetic resonance (NMR) spectroscopy, we monitor the kinetics of PbS solute generation, nucleation, and crystal growth from three thiourea precursors whose conversion reactivity spans a 2-fold range. In all three cases, nucleation is found to be slow and continues during >50% of the precipitation. A population balance model based on a size dependent growth law (1/r) fits the data with a single growth rate constant (k G) across all three precursors. However, the magnitude of the k G and the lack of solvent viscosity dependence indicates that the rate limiting step is not diffusion from solution to the nanoparticle surface. Several surface reaction limited mechanisms and a ligand penetration model that fits data our experiments using a single fit parameter are proposed to explain the results.

Growth kinetics determine the polydispersity and size of PbS and PbSe nanocrystals.

A library of thio- and selenourea derivatives is used to adjust the kinetics of PbE (E = S, Se) nanocrystal formation across a 1000-fold range (k r = 10-1 to 10-4 s-1), at several temperatures (80-120 °C), under a standard set of conditions (Pb : E = 1.2 : 1, [Pb(oleate)2] = 10.8 mM, [chalcogenourea] = 9.0 mM). An induction delay (t ind) is observed prior to the onset of nanocrystal absorption during which PbE solute is observed using in situ X-ray total scattering. Density functional theory models fit to the X-ray pair distribution function (PDF) support a Pb2(µ2-S)2(Pb(O2CR)2)2 structure. Absorption spectra of aliquots reveal a continuous increase in the number of nanocrystals over more than half of the total reaction time at low temperatures. A strong correlation between the width of the nucleation phase and reaction temperature is observed that does not correlate with the polydispersity. These findings are antithetical to the critical concentration dependence of nucleation that underpins the La Mer hypothesis and demonstrates that the duration of the nucleation period has a minor influence on the size distribution. The results can be explained by growth kinetics that are size dependent, more rapid at high temperature, and self limiting at low temperatures.

The darker side of head lice infestations.

Head lice (Pediculus humanus capitis) infestations are prevalent among young children and generally not considered a health hazard. Although massive chronic head lice infestations have been documented in paleo medical literature, their association with severe iron deficiency anemia has been rarely discussed in modern medicine. A recently published case implicating a head lice infestation as the cause of death of a 12-year-old girl brings this topic to the fore. This contribution discusses the available information concerning the circumstances under which this girl died and the ongoing criminal investigation as a springboard for discussion of how head lice infestations may serve as forensic and legal evidence and as a warning sign of mental illness, abuse, neglect, and rape.

Photophysical Properties of Cyclometalated Platinum(II) Diphosphine Compounds in the Solid State and in PMMA Films.

Platinum(II) compounds were synthesized with both chelate cyclometalated ligands and chelate diphosphine ligands. The cyclometalated ligands include phenylpyridine and a benzothiophene-containing ligand. The three new benzothiophene compounds were characterized by nuclear magnetic resonance (NMR) spectroscopy, high-resolution mass spectrometry (HR-MS), and photophysical measurements. In the case of one compound, L1-DPPM, the structure was determined by single crystal X-ray diffraction. The structural coherence of the noncrystalline emissive solid state was measured by X-ray total scattering real space pair distribution function analysis. Quantum yield values of all of the platinum compounds measured in the solid state and in PMMA films were much greater than in solution.

WgLink: reconstructing whole-genome viral haplotypes using L0+L1-regularization.

SUMMARY: Many tools can reconstruct viral sequences based on next-generation sequencing reads. Although existing tools effectively recover local regions, their accuracy suffers when reconstructing the whole viral genomes (strains). Moreover, they consume significant memory when the sequencing coverage is high or when the genome size is large. We present WgLink to meet this challenge. WgLink takes local reconstructions produced by other tools as input and patches the resulting segments together into coherent whole-genome strains. We accomplish this using an L0+L1-regularized regression, synthesizing variant allele frequency data with physical linkage between multiple variants spanning multiple regions simultaneously. WgLink achieves higher accuracy than existing tools both on simulated and on real datasets while using significantly less memory (RAM) and fewer CPU hours. AVAILABILITY AND IMPLEMENTATION: Source code and binaries are freely available at https://github.com/theLongLab/wglink. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Reconstruction of Microbial Haplotypes by Integration of Statistical and Physical Linkage in Scaffolding.

DNA sequencing technologies provide unprecedented opportunities to analyze within-host evolution of microorganism populations. Often, within-host populations are analyzed via pooled sequencing of the population, which contains multiple individuals or "haplotypes." However, current next-generation sequencing instruments, in conjunction with single-molecule barcoded linked-reads, cannot distinguish long haplotypes directly. Computational reconstruction of haplotypes from pooled sequencing has been attempted in virology, bacterial genomics, metagenomics, and human genetics, using algorithms based on either cross-host genetic sharing or within-host genomic reads. Here, we describe PoolHapX, a flexible computational approach that integrates information from both genetic sharing and genomic sequencing. We demonstrated that PoolHapX outperforms state-of-the-art tools tailored to specific organismal systems, and is robust to within-host evolution. Importantly, together with barcoded linked-reads, PoolHapX can infer whole-chromosome-scale haplotypes from 50 pools each containing 12 different haplotypes. By analyzing real data, we uncovered dynamic variations in the evolutionary processes of within-patient HIV populations previously unobserved in single position-based analysis.

kTWAS: integrating kernel machine with transcriptome-wide association studies improves statistical power and reveals novel genes.

The power of genotype-phenotype association mapping studies increases greatly when contributions from multiple variants in a focal region are meaningfully aggregated. Currently, there are two popular categories of variant aggregation methods. Transcriptome-wide association studies (TWAS) represent a set of emerging methods that select variants based on their effect on gene expressions, providing pretrained linear combinations of variants for downstream association mapping. In contrast to this, kernel methods such as sequence kernel association test (SKAT) model genotypic and phenotypic variance use various kernel functions that capture genetic similarity between subjects, allowing nonlinear effects to be included. From the perspective of machine learning, these two methods cover two complementary aspects of feature engineering: feature selection/pruning and feature aggregation. Thus far, no thorough comparison has been made between these categories, and no methods exist which incorporate the advantages of TWAS- and kernel-based methods. In this work, we developed a novel method called kernel-based TWAS (kTWAS) that applies TWAS-like feature selection to a SKAT-like kernel association test, combining the strengths of both approaches. Through extensive simulations, we demonstrate that kTWAS has higher power than TWAS and multiple SKAT-based protocols, and we identify novel disease-associated genes in Wellcome Trust Case Control Consortium genotyping array data and MSSNG (Autism) sequence data. The source code for kTWAS and our simulations are available in our GitHub repository (https://github.com/theLongLab/kTWAS).

Platinum Complexes from C-H Activation of Sterically Hindered [C

Primary amines and benzothiophene-3-carboxaldehyde were reacted to give four large, bulky imine ligands. These imine ligands were reacted with a tetramethyl platinum dimer and by heteroatom-assisted C-H activation, both monometalated compounds and bismetalated compounds were synthesized. In all cases, five-membered platinacycles were formed. The compounds were characterized by NMR spectroscopy, and one bismetalated compound was characterized by single-crystal X-ray diffraction. The UV-vis absorption and emission spectra and the excited-state lifetimes were recorded for these complexes. Density functional theory (DFT) and time-dependent-DFT calculations were performed to aid in the assignment of the absorption and emission spectra of the newly synthesized complexes.

Anthracene as a Launchpad for a Phosphinidene Sulfide and for Generation of a Phosphorus-Sulfur Material Having the Composition P2S, a Vulcanized Red Phosphorus That Is Yellow.

Thermolysis of a pair of dibenzo-7-phosphanorbornadiene compounds is shown to lead to differing behaviors: phosphinidene sulfide release and formation of amorphous P2S. These compounds, tBuP(S)A (1, A = C14H10 or anthracene; 59% isol. yield) and HP(S)A (2; 63%), are available through thionation of tBuPA and the new secondary phosphine HPA (5), prepared from Me2NPA and DIBAL-H in 50% yield. Phosphinidene sulfide [ tBuP═S] transfer is shown to proceed efficiently from 1 to 2,3-dimethyl-1,3-butadiene to form Diels-Alder product 3 with a zero-order dependence on diene. Platinum complex (Ph3P)2Pt(η2- tBuPS) (4, 47%) is also accessed from 1 and structurally characterized. In contrast, heating parent species 2 (3 h, 135 °C) under vacuum instead produces an insoluble, nonvolatile yellow residual material 6 of composition P2S that displays semiconductor properties with an optical band gap of 2.4 eV. Material 6 obtained in this manner from molecular precursor 2 is in a poorly characterized portion of the phosphorus-sulfur phase diagram and has therefore been subjected to a range of spectroscopic techniques to gain structural insight. X-ray spectroscopic and diffraction techniques, including Raman, XANES, EXAFS, and PDF, reveal 6 to have similarities with related compounds including P4S3, Hittorf's violet phosphorus. Various possible structures have been explored as well using quantum chemical calculations under the constraint that each phosphorus atom is trivalent with no terminal sulfide groups, and each sulfur atom is divalent. The structural conclusions are supported by data from phosphorus-31 magic angle spinning (MAS) solid state NMR spectroscopy, bolstering the structural comparisons to other phosphorus-sulfur systems while excluding the formulation of P2S as a simple mixture of P4S3 and phosphorus.

The Effect of Patient Characteristics and Sleep Quality on Visual Field Performance Reliability.

PURPOSE: To investigate the association of automated visual field (VF) reliability indices (false positive [FP], false negative [FN], and fixation loss [FL]) and sleep quality, VF experience, and age. METHODS: Prospective, cross-sectional study. Adult patients (age ≥ 18 years) completing automated VF testing were invited to participate. Baseline participant characteristics were obtained, and all participants were asked to complete the Pittsburgh Sleep Quality Index (PSQI) questionnaire. Nonparametric Spearman correlations and logistical regression models were performed. RESULTS: 63 patients were enrolled. Lower PSQI score was correlated with higher percentage (%) FL in the right eye (p = 0.03). Fewer prior VF was significantly correlated with higher %FP in the right eye (p = 0.008). Older age was significantly correlated with higher %FN in the left eye (p = 0.01). Greater mean deviation (MD) and pattern standard deviation (PSD) were strongly correlated with higher %FN in the right (p = 0.02 and 0.002, resp.) and left eyes (p = 0.01 and 0.02, resp.). CONCLUSION: In this prospective, cross-sectional study, worse MD and PSD are strongly correlated with increased FN in both eyes. Increased FN in the left eye associated with older age might be attributable to test fatigue. Worse sleep quality is associated with decreased FL in the right eye.

Ophthalmic findings in Frank-ter Haar syndrome: report of a sibling pair.

Frank-ter Haar syndrome (FTHS) is an autosomal recessive disorder characterized by abnormalities that affect the development of bone, heart, and eyes. We report a sibling pair with FTHS caused by a homozygous, novel mutation pLys133Glnfs*13 in the SH3PXD2B gene: one sibling had bilateral ocular hypertension and unilateral colobomas of iris, choroid and retina; the other, unilateral myelinated nerve fiber layer of the optic disk and papilledema due to idiopathic intracranial hypertension. Both children had refractive amblyopia and megalocornea.

Clinical management outcomes of childhood glaucoma suspects.

PURPOSE: To investigate the outcomes of childhood glaucoma suspects. DESIGN: Retrospective case series. METHODS: Records of childhood glaucoma suspects were identified using financial claims data; medical history, baseline biometric and exam findings were recorded. Conversion from suspect to glaucoma was determined based on the Childhood Glaucoma Research Network criteria. The study adheres to the tenets of the Declarations of Helsinki. RESULTS: 214 subjects were enrolled, with median age at initial presentation of 6.37 years (interquartertile range: Q1 = 2.46, Q3 = 8.90). 22 (10.2%) subjects developed glaucoma, 64 (29.9%) had ocular hypertension but no glaucoma, 9 (4.2%) had high-risk condition or syndrome without either ocular hypertension or glaucoma after a mean follow up of 39 +/- 34 months. Neither a family history of glaucoma nor patient gender was significantly different between the groups. 40.2% of subjects (86 of 214) had two or more episodes of intraocular pressure (IOP) > 21 mmHg, among which 25.6% (22 of 86) developed glaucoma after a mean duration of 32.8 +/- 33.5 months. CONCLUSIONS: Up to 25% of children with 2 or more episodes of elevated IOP may develop glaucoma. In 50% of suspects who converted to glaucoma, elevated IOP was not present at the initial evaluation. There is no significant difference in gender, family history, or baseline central corneal thickness between suspects who developed glaucoma compared to the rest. While suspects who converted to glaucoma had higher average, maximum and minimum IOP measurements, there is no clear cutoff between the groups.

Vision Loss Secondary to Meningioma in the Pregnant Patient.

A 31-year-old primagravid female at 27 weeks gestation presented to the emergency room with three weeks of progressive blurring of vision associated with intermittent headaches. Ocular examination revealed diminished visual acuity, decreased color discrimination, and constricted confrontation visual fields; optic nerve appearance was however normal. Magnetic resonance imaging of the brain and orbits revealed a large tuberculum meningioma compressing the optic chiasm and prechiasmatic optic nerves, as well as a small sphenoid wing meningioma. Given the risk of permanent vision loss, the patient underwent emergent tumor resection. Near total resection of the masses was achieved and the patient had complete resolution of her vision post-operatively. She gave birth via Caesarean section at 39 weeks. This case report describes the clinical presentations of intracranial meningiomas and discusses the challenges this condition poses in management during pregnancy.