Virtual Screening with Gnina 1.0

Virtual screening—predicting which compounds within a specified compound library bind to a target molecule, typically a protein—is a fundamental task in the field of drug discovery. Doing virtual screening well provides tangible practical benefits, including reduced drug development costs, faster time to therapeutic viability, and fewer unforeseen side effects. As with most applied computational tasks, the algorithms currently used to perform virtual screening feature inherent tradeoffs between speed and accuracy. Furthermore, even theoretically rigorous, computationally intensive methods may fail to account for important effects relevant to whether a given compound will ultimately be usable as a drug. Here we investigate the virtual screening performance of the recently released Gnina molecular docking software, which uses deep convolutional networks to score protein-ligand structures. We find, on average, that Gnina outperforms conventional empirical scoring. The default scoring in Gnina outperforms the empirical AutoDock Vina scoring function on 89 of the 117 targets of the DUD-E and LIT-PCBA virtual screening benchmarks with a median 1% early enrichment factor that is more than twice that of Vina. However, we also find that issues of bias linger in these sets, even when not used directly to train models, and this bias obfuscates to what extent machine learning models are achieving their performance through a sophisticated interpretation of molecular interactions versus fitting to non-informative simplistic property distributions.

Targeted Secondary Screening for Congenital Hypothyroidism in High-Risk Neonates: A 9 Year Review in a Large California Health Care System

Secondary screening for missed congenital hypothyroidism (CH) has been introduced sporadically, but its necessity and optimal strategy have not been recognized. We hypothesized that a simple clinical protocol (performed by a medical group without a governmental mandate) targeting infants at high risk for missed CH can identify cases. We performed a 9-year retrospective review of 338,478 neonates within a California health plan following the introduction of thyrotropin (TSH) secondary screening for neonates at high risk for missed CH due to very-low-birthweight (VLBW), hospitalized congenital heart disease (CHD), and same-sex multiples (SSM). Screening performance by day 60 of life was 95% successful for VLBW and >50% for CHD and SSM, leading to an additional 35% CH treated cases despite re-testing only 1.7% of the cohort. Infants with VLBW or CHD were 33 times more likely (190 times more likely for CHD with Down Syndrome) to receive treatment for CH than random infants diagnosed by primary screening (p < 0.001), and 92% of these infants were not found by primary newborn screening. Currently, permanent disease has been documented in 84% of CH by primary screening compared to 27% by secondary screening (p < 0.001). This targeted secondary screening program identifies and treats additional CH cases after TSH-only newborn screening.

Detecting of Proximal Caries in Primary Molars using Pen-type QLF Device

The purpose of this in vivo study was to assess the clinical screening performance of a quantitative light-induced fluorescence (QLF) device in detecting proximal caries in primary molars. Fluorescence loss, red autofluorescence and a simplified QLF score for proximal caries (QS-proximal) were evaluated for their validity in detecting proximal caries in primary molars compared to bitewing radiography. Three hundred and forty-four primary molar surfaces were included in the study. Carious lesions were scored according to lesion severity assessed by visual-tactile and radiographic examinations. The QLF images were analyzed for two quantitative parameters, fluorescence loss and red autofluorescence, as well as for QS-proximal. For both quantitative parameters and QS-proximal, the sensitivity, specificity and area under receiver operating curve (AUROC) were calculated as a function of the radiographic scoring index at enamel and dentin caries levels. Both quantitative parameters showed fair AUROC values for detecting dentine level caries (△F = 0.794, △R = 0.750). QS-proximal showed higher AUROC values (0.757 - 0.769) than that of visual-tactile scores (0.653) in detecting dentine level caries. The QLF device showed fair screening performance in detecting proximal caries in primary molars compared to bitewing radiography.

Validation of computer-aided diagnosis of diabetic retinopathy from retinal photographs of diabetic patients from Telecamps

Objective: To validate an algorithm previously developed by the Healthcare Technology Innovation Centre, IIT Madras, India for screening of diabetic retinopathy (DR), in fundus images of diabetic patients from telecamps to examine the screening performance for DR. Design: Photographs of patients with diabetes were examined using the automated algorithm for the detection of DR Setting: Mobile Teleophthalmology camps were conducted in two districts in Tamil Nadu, India from Jan 2015 to May 2017. Participants: 939 eyes of 472 diabetic patients were examined at Mobile Teleophthalmology camps in Thiruvallur and Kanchipuram districts, Tamil Nadu, India,. Fundus images were obtained (40-45-degree posterior pole in each eye) for all patients using a nonmydriatic fundus camera by the fundus photographer. Main Outcome Measures: Fundus images were evaluated for presence or absence of DR using a computer-assisted algorithm, by an ophthalmologist at a tertiary eye care centre (reference standard) and by a fundus photographer. Results: The algorithm demonstrated 85% sensitivity and 80% specificity in detecting DR compared to ophthalmologist. The area under the receiver operating characteristic curve was 0.69 (95%CI=0.65 to 0.73). The algorithm identified 100% of vision-threatening retinopathy just like the ophthalmologist. When compared to the photographer, the algorithm demonstrated 81% sensitivity and 78% specificity. The sensitivity of the photographer to detect DR was found to be 86% and specificity was 99% in detecting DR compared to ophthalmologist. Conclusions: The algorithm can detect the presence or absence of DR in diabetic patients. All findings of vision-threatening retinopathy could be detected with reasonable accuracy and will help to reduce the workload for human graders in remote areas.

Virtual Screening with Gnina 1.0

Virtual screening - predicting which compounds within a specified compound library bind to a target molecule, typically a protein - is a fundamental task in the field of drug discovery. Doing virtual screening well provides tangible practical benefits, including reduced drug development costs, faster time to therapeutic viability, and fewer unforeseen side effects. As with most applied computational tasks, the algorithms currently used to perform virtual screening feature inherent tradeoffs between speed and accuracy. Furthermore, even theoretically rigorous, computationally intensive methods may fail to account for important effects relevant to whether a given compound will ultimately be usable as a drug. Here we investigate the virtual screening performance of the recently released Gnina molecular docking software, which uses deep convolutional networks to score protein-ligand structures. We find, on average, that Gnina outperforms conventional empirical scoring. The default scoring in Gnina outperforms the empirical AutoDock Vina scoring function on 89 of the 117 targets of the DUD-E and LIT-PCBA virtual screening benchmarks with a median 1% early enrichment factor that is more than twice that of Vina. However, we also find that issues of bias linger in these sets, even when not used directly to train models, and this bias obfuscates to what extent machine learning models are achieving their performance through a sophisticated interpretation of molecular interactions versus fitting to non-informative simplistic property distributions.

Vibration Characteristics of an Industrial-Scale Flip-Flow Screen with Crank-Link Structure and Parameters Optimization

Flip-flow screens are increasingly used in the processing of fine wet coal. In this work, the vibration characteristics of an industrial-scale flip-flow screen with crank-link structure (FFSCLS) were investigated theoretically and experimentally. An improved kinematic model of FFSCLS was proposed and experiments are carried out to verify the reasonability. The effects of the key parameters of the eccentricity of the crankshaft, the rotational speed of the crankshaft, and the tension length of the screen surface on the vibration characteristics of the screen were investigated parametrically. The results show that the kinematic model can describe the vibration characteristics of screen perfectly with the maximum error between the theoretical and experimental results being within 6.96%. Moreover, the key parameters of the eccentricity of the crankshaft, the rotational speed of the crankshaft, and the tension length of the screen surface have significant effects on the vibrations of the screen body and screen surface. These parameters should be optimized to achieve maximum screening performance of the FFSCLS. This work should be useful for optimal design and efficient operation of the flip-flow screen.