Validation of machine-learning model for first-trimester prediction of pre-eclampsia using cohort from PREVAL study.

OBJECTIVE: Effective first-trimester screening for pre-eclampsia (PE) can be achieved using a competing-risks model that combines risk factors from the maternal history with multiples of the median (MoM) values of biomarkers. A new model using artificial intelligence through machine-learning methods has been shown to achieve similar screening performance without the need for conversion of raw data of biomarkers into MoM. This study aimed to investigate whether this model can be used across populations without specific adaptations. METHODS: Previously, a machine-learning model derived with the use of a fully connected neural network for first-trimester prediction of early (< 34 weeks), preterm (< 37 weeks) and all PE was developed and tested in a cohort of pregnant women in the UK. The model was based on maternal risk factors and mean arterial blood pressure (MAP), uterine artery pulsatility index (UtA-PI), placental growth factor (PlGF) and pregnancy-associated plasma protein-A (PAPP-A). In this study, the model was applied to a dataset of 10 110 singleton pregnancies examined in Spain who participated in the first-trimester PE validation (PREVAL) study, in which first-trimester screening for PE was carried out using the Fetal Medicine Foundation (FMF) competing-risks model. The performance of screening was assessed by examining the area under the receiver-operating-characteristics curve (AUC) and detection rate (DR) at a 10% screen-positive rate (SPR). These indices were compared with those derived from the application of the FMF competing-risks model. The performance of screening was poor if no adjustment was made for the analyzer used to measure PlGF, which was different in the UK and Spain. Therefore, adjustment for the analyzer used was performed using simple linear regression. RESULTS: The DRs at 10% SPR for early, preterm and all PE with the machine-learning model were 84.4% (95% CI, 67.2-94.7%), 77.8% (95% CI, 66.4-86.7%) and 55.7% (95% CI, 49.0-62.2%), respectively, with the corresponding AUCs of 0.920 (95% CI, 0.864-0.975), 0.913 (95% CI, 0.882-0.944) and 0.846 (95% CI, 0.820-0.872). This performance was achieved with the use of three of the biomarkers (MAP, UtA-PI and PlGF); inclusion of PAPP-A did not provide significant improvement in DR. The machine-learning model had similar performance to that achieved by the FMF competing-risks model (DR at 10% SPR, 82.7% (95% CI, 69.6-95.8%) for early PE, 72.7% (95% CI, 62.9-82.6%) for preterm PE and 55.1% (95% CI, 48.8-61.4%) for all PE) without requiring specific adaptations to the population. CONCLUSIONS: A machine-learning model for first-trimester prediction of PE based on a neural network provides effective screening for PE that can be applied in different populations. However, before doing so, it is essential to make adjustments for the analyzer used for biochemical testing. © 2023 International Society of Ultrasound in Obstetrics and Gynecology.

Machine-learning-based prediction of pre-eclampsia using first-trimester maternal characteristics and biomarkers.

OBJECTIVE: To evaluate the accuracy of predicting the risk of developing pre-eclampsia (PE) according to first-trimester maternal demographic characteristics, medical history and biomarkers using artificial-intelligence and machine-learning methods. METHODS: The data were derived from prospective non-interventional screening for PE at 11-13 weeks' gestation at two maternity hospitals in the UK. The data were divided into three subsets. The first set, including 30 437 subjects, was used to develop the training process, the second set of 10 000 subjects was utilized to optimize the machine-learning hyperparameters and the third set of 20 352 subjects was coded and used for model validation. An artificial neural network was used to predict from the demographic characteristics and medical history the prior risk that was then combined with biomarker values to determine the risk of PE and preterm PE with delivery at < 37 weeks' gestation. An additional network was trained without including race as input. Biomarkers included uterine artery pulsatility index (UtA-PI), mean arterial blood pressure (MAP), placental growth factor (PlGF) and pregnancy-associated plasma protein-A. All markers were entered using raw values without conversion into standardized multiples of the median. The prediction accuracy was estimated using the area under the receiver-operating-characteristics curve (AUC). We further computed the detection rate at 10%, 20% and 40% false-positive rates (FPR). The impact of taking aspirin was also added. Shapley values were calculated to evaluate the contribution of each parameter to the prediction of risk. We used a non-parametric test to compare the expected AUC with the one obtained when we randomly scrambled the labels and kept the predictions. For the general prediction, we performed 10 000 permutations of the labels. When the AUC was higher than the one obtained in all 10 000 permutations, we reported a P-value of < 0.0001. For the race-specific analysis, we performed 1000 permutations. When the AUC was higher than the AUC in permutations, we reported a P-value of < 0.001. RESULTS: The detection rate for preterm PE vs no PE, at a 10% FPR, was 53.3% when screening by maternal factors only, and the corresponding AUC was 0.816; these increased to 75.3% and 0.909, respectively, with the addition of biomarkers into the model. Information on race was important for the prediction accuracy; when race was not used to train the model, at a 10% FPR, the detection rate of preterm PE vs no PE decreased to 34.5-45.5% (for different races) when screening by maternal factors only and to 55.0-62.1% when biomarkers were added. The major predictors of PE were high MAP and UtA-PI, and low PlGF. The accuracy of prediction of all PE cases was lower than that for preterm PE. Aspirin use was recommended for cases who were at high risk of preterm PE. The AUC of all PE vs no PE was 0.770 when screening by maternal factors and 0.817 when the biomarkers were added; the respective detection rates, at a 10% FPR, were 41.3% and 52.9%. CONCLUSIONS: Screening for PE using a non-linear machine-learning-based approach does not require a population-based normalization, and its performance is similar to that of logistic regression. Removing race information from the model reduces its prediction accuracy, especially for the non-white populations when only maternal factors are considered. © 2022 International Society of Ultrasound in Obstetrics and Gynecology.

Invasion Rate Versus Diversity in Population Dynamics with Catastrophes.

A key question in the current diversity crisis is how diversity has been maintained throughout evolution and how to preserve it. Modern coexistence theories suggest that a high invasion rate of rare new types is directly related to diversity. We show that adding almost any mechanism of catastrophes to a stochastic birth, death, and mutation process with limited carrying capacity induces a novel phase transition characterized by a positive invasion rate but a low diversity. In this phase, new types emerge and grow rapidly, but the resulting growth of very large types decreases diversity. This model also resolves two major drawbacks of neutral evolution models: their failure to explain balancing selection without resorting to fitness differences and the unrealistic time required for the creation of the observed large types. We test this model on a classical case of genetic polymorphism: the HLA locus.

Fluctuations-induced coexistence in public goods dynamics.

Cooperative interactions between individuals in a population and their stability properties are central to population dynamics and evolution. We introduce a generic class of nonlinear dynamical systems describing such interactions between producers and non-producers of a rapidly equilibrating common resource extracted from a finite environment. In the deterministic mean field approximation, fast-growing non-producers drive the entire population to extinction. However, the presence of arbitrarily small perturbations destabilizes this fixed point into a stochastic attractor where both phenotypes can survive. Phase space arguments and moment closure are used to characterize the attractor and show that its properties are not determined by the noise amplitude or boundary conditions, but rather it is stabilized by the stochastic nonlinear dynamics. Spatial Monte Carlo simulations with demographic fluctuations and diffusion illustrate a similar effect, supporting the validity of the two-dimensional stochastic differential equation as an approximation. The functional distribution of the noise emerges as the main factor determining the dynamical outcome. Noise resulting from diffusion between different regions, or additive noise, induce coexistence while multiplicative or local demographic noise do not alter the outcome of deterministic dynamics. The results are discussed in a general context of the effect of noise on phase space structure.

Survival induced by short-scale autocorrelations in birth-death-diffusion models with negative growth.

In a birth, death, and diffusion process, the extinction-survival transition occurs when the average net growth rate is zero. For instance, in the presence of normally distributed time-varying stochastic growth rates with no autocorrelation, the transition indeed occurs at zero net growth rates. In contrast, if the growth rates are constant in time, a large enough variance in the growth rate will systematically ensure the survival of the global population even in a small system and, more importantly, even with a negative net growth rate. We here show that, surprisingly, for any intermediate temporal autocorrelation, any length of correlation, and any negative average growth rate, the same result holds. We test this argument on exponential and power law autocorrelation models and propose a simple condition for the growth rate variance at the transition.

High fraction of silent recombination in a finite-population two-locus neutral birth-death-mutation model.

A precise estimate of allele and haplotype polymorphism is of great interest in theoretical population genetics, but also has practical applications, such as bone marrow registries management. Allele polymorphism is driven mainly by point mutations, while haplotype polymorphism is also affected by recombination. Current estimates treat recombination as mutations in an infinite site model. We here show that even in the simple case of two loci in a haploid individual, for a finite population, most recombination events produce existing haplotypes, and as such are silent. Silent recombination considerably reduces the total number of haplotypes expected from the infinite site model for populations that are not much larger than one over the mutation rate. Moreover, in contrast with mutations, the number of haplotypes does not grow linearly with the population size. We hence propose a more accurate estimate of the total number of haplotypes that takes into account silent recombination. We study large-scale human leukocyte antigen (HLA) haplotype frequencies from human populations to show that the current estimated recombination rate in the HLA region is underestimated.

Family-size variability grows with collapse rate in a birth-death-catastrophe model.

Forest-fire and avalanche models support the notion that frequent catastrophes prevent the growth of very large populations and as such, prevent rare large-scale catastrophes. We show that this notion is not universal. A new model class leads to a paradigm shift in the influence of catastrophes on the family-size distribution of subpopulations. We study a simple population dynamics model where individuals, as well as a whole family, may die with a constant probability, accompanied by a logistic population growth model. We compute the characteristics of the family-size distribution in steady state and the phase diagram of the steady-state distribution and show that the family and catastrophe size variances increase with the catastrophe frequency, which is the opposite of common intuition. Frequent catastrophes are balanced by a larger net-growth rate in surviving families, leading to the exponential growth of these families. When the catastrophe rate is further increased, a second phase transition to extinction occurs when the rate of new family creations is lower than their destruction rate by catastrophes.

Adaptation of autocatalytic fluctuations to diffusive noise.

Evolution of a system of diffusing and proliferating mortal reactants is analyzed in the presence of randomly moving catalysts. While the continuum description of the problem predicts reactant extinction as the average growth rate becomes negative, growth rate fluctuations induced by the discrete nature of the agents are shown to allow for an active phase, where reactants proliferate as their spatial configuration adapts to the fluctuations of the catalyst density. The model is explored by employing field theoretical techniques, numerical simulations, and strong coupling analysis. For d< or =2, the system is shown to exhibits an active phase at any growth rate, while for d>2 a kinetic phase transition is predicted. The applicability of this model as a prototype for a host of phenomena that exhibit self-organization is discussed.

Cell surface dynamics: the balance between diffusion, aggregation and endocytosis.

The cell membrane lies at the interface between an extracellular set of signals and the appropriate intracellular response. Specifically, lymphocyte activity is determined by the spatial and structural response to antigens, as mediated by cell surface receptors. In order to correlate experimentally observed cellular activities, such as secretion, anergy, death, survival and division to external stimuli, it is necessary to monitor cell surface dynamics. B-lymphocyte activation results from the stimulation by large immune complexes comprising antigens, B-cell receptors (BcRs) and co-receptors. Compartmentalisation of the interacting molecular components is required in order to assure the rapid initiation of specialised and sustained signalling cascades. In this study, a Monte Carlo simulation of the cell membrane dynamics was developed to clarify the receptor dynamics, aggregation mechanisms and their combined effect on cellular functions. This simulation is based on experimentally measured parameters and represents a feasible, advanced and reliable framework to investigate the cell surface. The current study focussed on B-cell surface dynamics. A model demonstrating the basic properties of BcR dynamics and how BcR kinetics is affected by lipid rafts is developed. The authors studied BcR interactions with multivalent ligands and the influence of lipid rafts on this interaction. Finally, the dynamics of the initial steps of BcR-mediated cell activation is estimated and the effect of the association of signalling molecules with lipid rafts is demonstrated. These results are used to suggest some novel hypotheses on BcR-mediated B-cell activation.

The role of lipid rafts in cell surface BCR antigen binding dynamics.

B lymphocyte activation results from the stimulation by large immune complexes involving antigens, antibodies, rafts and complement factors. Cell activation requires co-localization of the interacting molecular components. One of the main elements leading to this localization is the presence on the cell surface of lipid rafts. We show here that an appropriate amount of lipid rafts help to significantly (2- 3 orders of magnitude) raise the sensitivity of B lymphocyte to surrounding high valence antigens. The analysis was done using a newly developed graphically visualized, Monte Carlo (MC) simulation of the cell surface dynamics. Currently this platform represents a feasible, advanced and reliable framework to investigate the cell surface in general. We describe the model and determine, utilizing our model, the effect of lipid rafts surface fraction on the properties of B cell response to immune complexes. We validate our results using an approximate set of ODEs.

Modeling the influence of TH1- and TH2-type cells in autoimmune diseases.

A sharp TH1/TH2 dichotomy has often been used to define the effects of cytokines on autoimmune diseases. However contradictory results in recent research indicate that the situation may be more complex. Here, we build a simple mathematical model aimed at settling the contradictions. The model is applied using Ordinary Differential Equations (ODE). We show here that a TH1/TH2 paradigm is only an external view of a complex multivariate system.

The importance of being discrete: life always wins on the surface.

Many systems in chemistry, biology, finance, and social sciences present emerging features that are not easy to guess from the elementary interactions of their microscopic individual components. In the past, the macroscopic behavior of such systems was modeled by assuming that the collective dynamics of microscopic components can be effectively described collectively by equations acting on spatially continuous density distributions. It turns out that, to the contrary, taking into account the actual individual/discrete character of the microscopic components of these systems is crucial for explaining their macroscopic behavior. In fact, we find that in conditions in which the continuum approach would predict the extinction of all of the population (respectively the vanishing of the invested capital or the concentration of a chemical substance, etc.), the microscopic granularity insures the emergence of macroscopic localized subpopulations with collective adaptive properties that allow their survival and development. In particular it is found that in two dimensions "life" (the localized proliferating phase) always prevails.