Explore the vast range of titles available.

Accurate prediction of human pharmacokinetics (PK) for lead compounds is one of the critical determinants of successful drug development. Traditional methods for PK parameter prediction, such as in vitro to in vivo extrapolation and physiologically based pharmacokinetic modeling, often require extensive experimental data and time‐consuming calibration of parameters. Machine learning (ML) has been widely applied to predict ADME and physicochemical properties (ADMEP descriptors), but studies focusing on concentration‐time (C‐t) profile prediction remain limited. In this study, we developed a Long Short‐Term Memory (LSTM) based ML framework to predict C‐t profiles following intravenous (IV) bolus drug administration in humans. The model used ADMEP descriptors generated by ADMETlab 3.0 and dose information as input. A total of 40 drugs were used for training and 18 for testing, with concentration data simulated from published PK models. Our approach achieved R2 of 0.75 across all C‐t profiles, and 77.8% of Cmax, 55.6% of clearance, and 61.1% of volume of distribution predictions within a 2‐fold error range, demonstrating predictive performance comparable to previously published ML methods. Furthermore, model performance was found to be associated with the input dose level and ADMEP descriptors, suggesting the accuracy and confidence of the prediction may be expected in advance via these descriptors. This LSTM‐based framework using a small number of compounds enables efficient prediction of human PK profiles with IV dosing, offering a practical alternative to traditional PK prediction models. It holds promise for improving early‐phase prioritizing lead compounds and reducing reliance on animals in drug development.

Predicting pharmacokinetic (PK) profiles from molecular structures represents a significant advancement in pharmaceutical research with substantial implications for expediting drug discovery processes. We evaluated five approaches to systematically compare five distinct methodological frameworks for predicting rat plasma concentration‐time profiles directly from molecular structures using a consistent dataset and evaluation framework: (1) NCA‐ML, predicted non‐compartmental analysis parameters with one compartmental PK modeling; (2) PBPK‐ML, utilizing ML‐predicted in vitro characteristics in physiologically based PK models; (3) CMT‐ML, neural networks predicting compartmental PK model parameters with two or three compartmental PK modeling; (4) CMT‐PINN, employing physics‐informed neural networks trained on concentration‐time profiles predicting compartmental PK model parameters with two or three compartmental PK modeling; and (5) PURE‐ML, using decision trees to predict concentration values at specific time points. The CMT‐PINN approach achieved highest predictive performance closely followed by PURE‐ML (R2‐log: 0.854 vs. 0.789, Spearman: 0.933 vs. 0.896), with 65.9% versus 61.0% of predictions within a twofold error of the observed concentrations. The other three approaches showed substantially lower performance metrices and higher prediction error margins. Models trained directly on concentration‐time profiles outperformed those trained using derived PK parameters, particularly with limited training datasets. Our findings confirm the viability of predicting PK behavior from molecular structures prior to synthesis. The implementation of these computational approaches enables informed compound selection early in discovery, concentrating resources on promising candidates, and potentially reducing animal studies while accelerating development timelines. Study Highlights What is the current knowledge on the topic? ○Various machine learning approaches predict PK profiles from molecular structures; however, these methods differ significantly in methodology and datasets, making direct comparisons challenging without a standardized evaluation framework. What question did this study address? ○Which machine‐learning approach most accurately predicts in vivo rat plasma concentration‐time profiles from molecular structures when evaluated using identical datasets, consistent data partitioning, and uniform performance metrics? What does this study add to our knowledge? ○Physics‐informed neural networks (CMT‐PINN) trained directly on concentration‐time profiles outperformed other approaches, achieving superior accuracy, even with smaller datasets and a more compact neural network architecture. How might this change drug discovery, development, and/or therapeutics? ○Implementation of superior PK prediction methods enables earlier informed compound selection, reduces animal studies, accelerates research timelines, and advances the “in silico first” paradigm for more efficient drug discovery.

Background It has been reported that the protein binding rate of vancomycin (VCM) varies among individual patients. So, the authors investigated relevant factors that may affect free VCM concentration and target attainment of free area under the concentration-time curve (fAUC). Methods Thirty-nine patients were included. Multiple regression analysis was performed to determine the valuable factors in the free VCM concentration, and the target attainment of area under the concentration-time curve (AUC) 400–600 mg・h/L and fAUC200-300 mg・h/L was calculated. Results We found total protein was significant covariate for free VCM. Among 18 patients who were investigated for AUC and fAUC estimation, 9 patients (50.0%) and 12 patients (66.7%) reached AUC > 600 mg・h/L, and fAUC > 300 mg・h/L ( p  = 0.310), respectively. Conclusions Total protein is a significant predictor for free VCM estimation. And the fAUC-guided TDM for VCM TDM may contribute to more strict dosing than the AUC-guided TDM in hyper- or hypo-proteinemic population. Trial registration Retrospectively registered.

To assess the pharmacokinetic parameters of vancomycin in Chinese critically ill pediatric patients, children treated with vancomycin, hospitalized in the intensive care unit were included. Samples to determine peak and trough serum concentrations were obtained on the third day of treatment. Half‐life was significantly longer in neonates and showed a decreasing trend in infants and children. In patients aged ≥1 month, AUC24/MIC ≥400 was achieved in 31.8% at the dose of 40 mg/kg/d, and in 48.7% at the dose of 60 mg/kg/d with an assumed MIC of 1 mg/L. Augmented renal clearance (ARC) was present in 27.3% of children, which was associated with higher vancomycin clearance and lower AUC values. A good correlation was observed between trough concentration and AUC24, and the trough concentration that correlated with AUC24 of 400 were varied according to the dosage regimens, 8.42 mg/L for 6‐hintervals, and 6.63 mg/L for 8‐h intervals. To conclude, vancomycin trough concentration that related to the AUC24 of 400 was much lower in critically ill children than that in adults. The dosage of 60 mg/kg/day did not enough for producing AUC24 in the range of 400–600 mg h/L in critically ill children, especially in those with ARC. A good correlation was observed between trough concentration and AUC24. The trough concentration that correlated with AUC24 of 400 were varied according to the dosage regimens, 8.42 mg/L for 6‐h intervals, and 6.63 mg/L for 8‐h intervals.

Event-based estimates of the design flood in ungauged catchments are normally based on a single catchment response time parameter expressed as either the time of concentration (TC), lag time (TL) and/or time to peak (TP). In small, gauged catchments, a simplified convolution process between a single observed hyetograph and hydrograph is generally used to estimate these time parameters. In medium to large heterogeneous, gauged catchments, such a simplification is neither practical nor applicable, given that the variable antecedent soil moisture status resulting from previous rainfall events and spatially non-uniform rainfall hyetographs can result in multi-peaked hydrographs. In ungauged catchments, time parameters are estimated using either empirical or hydraulic methods. In South Africa (SA), unfortunately, the majority of the empirical methods recommended for general use were developed and verified in catchments ≤ 0.45 km² without using any local data. This paper presents the further development and verification of the streamflow-based approach developed by Gericke (2016) to estimate observed TP values and to derive a regional empirical TP equation in Primary Drainage Region X, SA. A semi-automated hydrograph analysis tool was developed to extract and analyse complete hydrographs for time parameter estimation using primary streamflow data from 51 flowgauging sites. The observed TP values were estimated using three methods: (i) duration of total net rise of a multi-peaked hydrograph, (ii) triangular-shaped direct runof hydrograph approximations, and (iii) linear catchment response functions. The combined use of these methods incorporated the high variability of eventbased time parameters, and Method (iii), in conjunction with an ensemble-event approach sampled from the time parameter distributions, should replace the event-based approaches to enable the improved calibration of empirical time parameter equations. The conceptual approach used to derive the regional empirical TP equation should also be adopted when regional equations need to be derived at a national scale in SA.

In this work the attenuation potential of flood peaks of 265 large reservoirs all over Italy is analysed, considering a flood management that excludes gates opening and then configures strictly unsupervised attenuation effects. Key factors of dams and related basins are considered to develop a ranking method that can emphasize the interplay between dam geometry and the hydrological processes acting in the upstream watershed. To maintain a homogeneous approach in such a wide geographic area, the attenuation index is computed applying the numerical solution of the differential equation of lakes and only two different standardized hydrograph shapes have been used. An index design flood from the rational method is used as the incoming peak value for each dam, enhancing the use of the results of a recent analysis of all Italian rainfall extremes. Even with a very simple approach, twenty-four different design incoming floods are derived, by varying the shape of the incoming hydrograph and the parameters of the rational method. Exploring the ranking results in all the alternatives, the attenuation potential obtained for all dams demonstrates to be strongly sensitive to the assumptions on the time of concentration and to some rainfall features. On the other hand, the hydrograph shape seems to exert much less influence on the ranking outcome. Results obtained can be useful to studies of wide-area flood frequency analyses, as we highlighted the sensitivity of the rank of attenuation efficiency to hydrologic parameters widely used in the assessment of the design flood peaks in ungauged basins.

Time of concentration (Tc) is a crucial aspect in the hydrological model, especially for determining flood discharge. Time concentration is the amount of time required for water to go from a watershed’s farthest point to its outflow. Although it is an important value, however, Tc does not have a universal equation that can be used as a reference. Based on the literature review, there are several concentration time calculation methods. Each method has its parameters and approach to produce varied values. Several experimental methods were applied to calculate concentration time at sites, among others are Kirpich, SCS Lag, and FAA. The calculation results are displayed in a dendrogram with group division by using the Euclidean approach. The calculation results show that the difference in Tc values can reach up to 500%, hence Grimaldi calls it a paradox in modern hydrology.

The therapeutic drug monitoring (TDM) of vancomycin (VCM) in critically ill patients often results in the estimated area being under the concentration–time curve (AUC) values that deviate from individual observations. In this study, we investigated the factors influencing the achievement of the target AUC of VCM at steady-state in critically ill patients. We retrospectively collected data from patients treated with VCM in an intensive care unit (ICU). Multivariate analysis was used to adjust for significant factors with p < 0.05 and identify new factors affecting the achievement of the target AUC at steady-state for VCM. Among the 113 patients included in this study, 72 (64%) were in the 1-point group (trough only), whereas 41 (36%) were in the 2-point group (trough/peak). The percentage of patients achieving the target AUC at the follow-up TDM evaluation was significantly higher in the two-point group. Multivariate analysis showed that being in the 2-point group and those with a 20% or more increase (or decrease) in creatinine clearance (CCr) were both significantly associated with the success rate of achieving the target AUC at the follow-up TDM. Novel findings revealed that in patients admitted to the ICU, changes in renal function were a predictor of AUC deviation, with a 20% or more increase (or decrease) in CCr being an indicator. We believe the indicators obtained in this study are simple and can be applied clinically in many facilities. If changes in renal function are anticipated, we recommend an AUC evaluation of VCM with a two-point blood collection, close monitoring of renal function, and dose adjustment based on reanalyzing the serum concentrations of VCM.

The Asian citrus psyllid, Diaphorina citri Kuwayama (Hemiptera: Liviidae), is a major pest of citrus crops worldwide. A large number of insecticides have been used to manage D. citri in Florida. Therefore, insecticide resistance could become an important problem facing citrus production. Monitoring insecticide susceptibility in populations of D. citri and providing a technique to use as an early warning is needed so citrus producers can modify chemical control strategies for this pest in Florida. The objective of this study was to develop a simple and fast tool to determine insecticide resistance in D. citri and apply it to commercial citrus production in Florida. LC50 and LC95 estimates were determined for 8 commonly used insecticides on a susceptible laboratory population of D. citri 24 h after treatment in a residual contact bottle assay. Five to 7 concentrations of each insecticide were tested. The LC50 values (and 95% fiducial limits) ranged from 0.06 (0.02–0.26) to 0.80 (0.26–2.46) ng/μL for each insecticide tested. Exposure time—mortality indices were determined for 0, 10, 100, 1,000, and 10,000 ng/μL concentrations of each insecticide in a laboratory susceptible strain. Knockdown was assessed after 15, 30, 45, 60, 75, 90, 105, and 120 min. Complete knockdown (100.0%) occurred within 60 min for dimethoate, fenpropathrin, imidacloprid, bifenthrin, and flupyradifurone at the 10,000 ng/μL concentration. For spinetoram, 86.7% knockdown occurred within 120 min at 10,000 ng/μL. For sulfoxaflor and cyantraniliprole, 44.0 and 42.6% knockdown, respectively, occurred within 120 min at 1,000 ng/μL. We also developed a bottle bioassay to survey field populations of D. citri for insecticide resistance in central Florida. Exposure time—mortality indices developed in the laboratory were used to assess susceptibility of 1 laboratory and 4 field populations of D. citri after 15, 30, 50, 75, 90, 105, and 120 min of exposure at the 10,000 ng/μL concentration of various insecticides. Little to no evidence of resistance was detected for bifenthrin, dimethoate, imidacloprid, and fenpropathrin in central Florida. Our investigation demonstrated that a bottle bioassay is suitable for assaying insecticide resistance in D. citri adults under laboratory and field conditions. It should be a flexible tool for rapid testing of insecticide resistance in possible cases of insecticide failure. Its simplicity should allow trained professionals to rapidly monitor for insecticide resistance in commercial settings where “hot spots” of D. citri populations may occur.

Due to the steep slope of mountainous watersheds and large changes in vegetation coverage degree, flood response processes after rainstorms are complicated. The flow concentration time of the slope is a key parameter for the simulation of flood processes. The most widely used flow concentration time formula currently in the distributed hydrological model is T = L0.6n0.6i−0.4S−0.3, which is derived from the kinematic wave theory (Melesse and Graham in J Am Water Resour As 40(4):863–879, 2004; Lee in Hydrol Sci 53(2):323–337, 2008). The flow confluence time T is characterized by the constant exponent of the slope length L, roughness n, effective rainfall intensity i and slope S, and the influence of vegetation on the flow concentration time is implied by the roughness. In this study, a series of heavy rainfall slope surface confluence tests under different slopes and vegetation coverage were carried out, a vegetation coverage factor, C, which was introduced, a statistical analysis method was used, and the vegetation coverage index was fitted. The results showed that the types of vegetation have a certain influence on the flow concentration time of slope, and the flow confluence time under turf vegetation was larger than the flow confluence time under shrubs vegetation; especially in the slope of the larger slope, the relative impact is more significant; at the same time, the influence of vegetation coverage on the flow concentration time of slope was more significant; no matter the condition of turf or shrub, the slope confluence time increased obviously with the increase in vegetation coverage. The index of vegetation coverage factor C varied with the slope and rain intensity. In general, the index of vegetation coverage factor C increased with the decrease in slope and decreased with the increase in rain intensity. In regard to the turf vegetation coverage index, when the slope is 45° and 30°, the decreasing trend of the vegetation coverage index a0 is obvious with increasing rainfall intensity. When the slope is 15°, the vegetation coverage index a0 also decreases with increasing rainfall intensity. When the slope is 5°, the vegetation coverage index a0 basically has no change. In regard to the shrubs vegetation coverage index, when the slope is 45° and 30°, the decreasing trend of the vegetation coverage index a0 is obvious with increasing rainfall intensity. When the slope is 15°, the vegetation coverage index a0 also decreases with increasing rainfall intensity. When the slope is 5°, the vegetation coverage index a0 basically has no change.