Surrogate science: The idol of a universal method for scientific inference

The application of statistics to science is not a neutral act. Statistical tools have shaped and were also shaped by its objects. In the social sciences, statistical methods fundamentally changed research practice, making statistical inference its centerpiece. At the same time, textbook writers in the social sciences have transformed rivaling statistical systems into an apparently monolithic method that could be used mechanically. The idol of a universal method for scientific inference has been worshipped since the "inference revolution" of the 1950s. Because no such method has ever been found, surrogates have been created, most notably the quest for significant p values. This form of surrogate science fosters delusions and borderline cheating and has done much harm, creating, for one, a flood of irreproducible results. Proponents of the "Bayesian revolution" should be wary of chasing yet another chimera: an apparently universal inference procedure. A better path would be to promote both an understanding of the various devices in the "statistical toolbox" and informed judgment to select among these.

Prevalence of iodine inadequacy in Switzerland assessed by the estimated average requirement cut-point method in relation to the impact of iodized salt.

OBJECTIVE: To assess the iodine status of Swiss population groups and to evaluate the influence of iodized salt as a vector for iodine fortification. DESIGN: The relationship between 24 h urinary iodine and Na excretions was assessed in the general population after correcting for confounders. Single-day intakes were estimated assuming that 92 % of dietary iodine was excreted in 24 h urine. Usual intake distributions were derived for male and female population groups after adjustment for within-subject variability. The estimated average requirement (EAR) cut-point method was applied as guidance to assess the inadequacy of the iodine supply. SETTING: Public health strategies to reduce the dietary salt intake in the general population may affect its iodine supply. SUBJECTS: The study population (1481 volunteers, aged ≥15 years) was randomly selected from three different linguistic regions of Switzerland. RESULTS: The 24 h urine samples from 1420 participants were determined to be properly collected. Mean iodine intakes obtained for men (n 705) and women (n 715) were 179 (sd 68.1) µg/d and 138 (sd 57.8) µg/d, respectively. Urinary Na and Ca, and BMI were significantly and positively associated with higher iodine intake, as were men and non-smokers. Fifty-four per cent of the total iodine intake originated from iodized salt. The prevalence of inadequate iodine intake as estimated by the EAR cut-point method was 2 % for men and 14 % for women. CONCLUSIONS: The estimated prevalence of inadequate iodine intake was within the optimal target range of 2-3 % for men, but not for women.

Error models in hydrogeology applications

Notre consommation en eau souterraine, en particulier comme eau potable ou pour l'irrigation, a considérablement augmenté au cours des années. De nombreux problèmes font alors leur apparition, allant de la prospection de nouvelles ressources à la remédiation des aquifères pollués. Indépendamment du problème hydrogéologique considéré, le principal défi reste la caractérisation des propriétés du sous-sol. Une approche stochastique est alors nécessaire afin de représenter cette incertitude en considérant de multiples scénarios géologiques et en générant un grand nombre de réalisations géostatistiques. Nous rencontrons alors la principale limitation de ces approches qui est le coût de calcul dû à la simulation des processus d'écoulements complexes pour chacune de ces réalisations. Dans la première partie de la thèse, ce problème est investigué dans le contexte de propagation de l'incertitude, oú un ensemble de réalisations est identifié comme représentant les propriétés du sous-sol. Afin de propager cette incertitude à la quantité d'intérêt tout en limitant le coût de calcul, les méthodes actuelles font appel à des modèles d'écoulement approximés. Cela permet l'identification d'un sous-ensemble de réalisations représentant la variabilité de l'ensemble initial. Le modèle complexe d'écoulement est alors évalué uniquement pour ce sousensemble, et, sur la base de ces réponses complexes, l'inférence est faite. Notre objectif est d'améliorer la performance de cette approche en utilisant toute l'information à disposition. Pour cela, le sous-ensemble de réponses approximées et exactes est utilisé afin de construire un modèle d'erreur, qui sert ensuite à corriger le reste des réponses approximées et prédire la réponse du modèle complexe. Cette méthode permet de maximiser l'utilisation de l'information à disposition sans augmentation perceptible du temps de calcul. La propagation de l'incertitude est alors plus précise et plus robuste. La stratégie explorée dans le premier chapitre consiste à apprendre d'un sous-ensemble de réalisations la relation entre les modèles d'écoulement approximé et complexe. Dans la seconde partie de la thèse, cette méthodologie est formalisée mathématiquement en introduisant un modèle de régression entre les réponses fonctionnelles. Comme ce problème est mal posé, il est nécessaire d'en réduire la dimensionnalité. Dans cette optique, l'innovation du travail présenté provient de l'utilisation de l'analyse en composantes principales fonctionnelles (ACPF), qui non seulement effectue la réduction de dimensionnalités tout en maximisant l'information retenue, mais permet aussi de diagnostiquer la qualité du modèle d'erreur dans cet espace fonctionnel. La méthodologie proposée est appliquée à un problème de pollution par une phase liquide nonaqueuse et les résultats obtenus montrent que le modèle d'erreur permet une forte réduction du temps de calcul tout en estimant correctement l'incertitude. De plus, pour chaque réponse approximée, une prédiction de la réponse complexe est fournie par le modèle d'erreur. Le concept de modèle d'erreur fonctionnel est donc pertinent pour la propagation de l'incertitude, mais aussi pour les problèmes d'inférence bayésienne. Les méthodes de Monte Carlo par chaîne de Markov (MCMC) sont les algorithmes les plus communément utilisés afin de générer des réalisations géostatistiques en accord avec les observations. Cependant, ces méthodes souffrent d'un taux d'acceptation très bas pour les problèmes de grande dimensionnalité, résultant en un grand nombre de simulations d'écoulement gaspillées. Une approche en deux temps, le "MCMC en deux étapes", a été introduite afin d'éviter les simulations du modèle complexe inutiles par une évaluation préliminaire de la réalisation. Dans la troisième partie de la thèse, le modèle d'écoulement approximé couplé à un modèle d'erreur sert d'évaluation préliminaire pour le "MCMC en deux étapes". Nous démontrons une augmentation du taux d'acceptation par un facteur de 1.5 à 3 en comparaison avec une implémentation classique de MCMC. Une question reste sans réponse : comment choisir la taille de l'ensemble d'entrainement et comment identifier les réalisations permettant d'optimiser la construction du modèle d'erreur. Cela requiert une stratégie itérative afin que, à chaque nouvelle simulation d'écoulement, le modèle d'erreur soit amélioré en incorporant les nouvelles informations. Ceci est développé dans la quatrième partie de la thèse, oú cette méthodologie est appliquée à un problème d'intrusion saline dans un aquifère côtier. -- Our consumption of groundwater, in particular as drinking water and for irrigation, has considerably increased over the years and groundwater is becoming an increasingly scarce and endangered resource. Nofadays, we are facing many problems ranging from water prospection to sustainable management and remediation of polluted aquifers. Independently of the hydrogeological problem, the main challenge remains dealing with the incomplete knofledge of the underground properties. Stochastic approaches have been developed to represent this uncertainty by considering multiple geological scenarios and generating a large number of realizations. The main limitation of this approach is the computational cost associated with performing complex of simulations in each realization. In the first part of the thesis, we explore this issue in the context of uncertainty propagation, where an ensemble of geostatistical realizations is identified as representative of the subsurface uncertainty. To propagate this lack of knofledge to the quantity of interest (e.g., the concentration of pollutant in extracted water), it is necessary to evaluate the of response of each realization. Due to computational constraints, state-of-the-art methods make use of approximate of simulation, to identify a subset of realizations that represents the variability of the ensemble. The complex and computationally heavy of model is then run for this subset based on which inference is made. Our objective is to increase the performance of this approach by using all of the available information and not solely the subset of exact responses. Two error models are proposed to correct the approximate responses follofing a machine learning approach. For the subset identified by a classical approach (here the distance kernel method) both the approximate and the exact responses are knofn. This information is used to construct an error model and correct the ensemble of approximate responses to predict the "expected" responses of the exact model. The proposed methodology makes use of all the available information without perceptible additional computational costs and leads to an increase in accuracy and robustness of the uncertainty propagation. The strategy explored in the first chapter consists in learning from a subset of realizations the relationship between proxy and exact curves. In the second part of this thesis, the strategy is formalized in a rigorous mathematical framework by defining a regression model between functions. As this problem is ill-posed, it is necessary to reduce its dimensionality. The novelty of the work comes from the use of functional principal component analysis (FPCA), which not only performs the dimensionality reduction while maximizing the retained information, but also allofs a diagnostic of the quality of the error model in the functional space. The proposed methodology is applied to a pollution problem by a non-aqueous phase-liquid. The error model allofs a strong reduction of the computational cost while providing a good estimate of the uncertainty. The individual correction of the proxy response by the error model leads to an excellent prediction of the exact response, opening the door to many applications. The concept of functional error model is useful not only in the context of uncertainty propagation, but also, and maybe even more so, to perform Bayesian inference. Monte Carlo Markov Chain (MCMC) algorithms are the most common choice to ensure that the generated realizations are sampled in accordance with the observations. Hofever, this approach suffers from lof acceptance rate in high dimensional problems, resulting in a large number of wasted of simulations. This led to the introduction of two-stage MCMC, where the computational cost is decreased by avoiding unnecessary simulation of the exact of thanks to a preliminary evaluation of the proposal. In the third part of the thesis, a proxy is coupled to an error model to provide an approximate response for the two-stage MCMC set-up. We demonstrate an increase in acceptance rate by a factor three with respect to one-stage MCMC results. An open question remains: hof do we choose the size of the learning set and identify the realizations to optimize the construction of the error model. This requires devising an iterative strategy to construct the error model, such that, as new of simulations are performed, the error model is iteratively improved by incorporating the new information. This is discussed in the fourth part of the thesis, in which we apply this methodology to a problem of saline intrusion in a coastal aquifer.

CNDOL: A fast and reliable method for the calculation of electronic properties of very large systems. Applications to retinal binding pocket in rhodopsin and gas phase porphine.

Very large molecular systems can be calculated with the so called CNDOL approximate Hamiltonians that have been developed by avoiding oversimplifications and only using a priori parameters and formulas from the simpler NDO methods. A new diagonal monoelectronic term named CNDOL/21 shows great consistency and easier SCF convergence when used together with an appropriate function for charge repulsion energies that is derived from traditional formulas. It is possible to obtain a priori molecular orbitals and electron excitation properties after the configuration interaction of single excited determinants with reliability, maintaining interpretative possibilities even being a simplified Hamiltonian. Tests with some unequivocal gas phase maxima of simple molecules (benzene, furfural, acetaldehyde, hexyl alcohol, methyl amine, 2,5 dimethyl 2,4 hexadiene, and ethyl sulfide) ratify the general quality of this approach in comparison with other methods. The calculation of large systems as porphine in gas phase and a model of the complete retinal binding pocket in rhodopsin with 622 basis functions on 280 atoms at the quantum mechanical level show reliability leading to a resulting first allowed transition in 483 nm, very similar to the known experimental value of 500 nm of "dark state." In this very important case, our model gives a central role in this excitation to a charge transfer from the neighboring Glu(-) counterion to the retinaldehyde polyene chain. Tests with gas phase maxima of some important molecules corroborate the reliability of CNDOL/2 Hamiltonians.

High c-Met expression in stage I-II pancreatic adenocarcinoma: proposal for an immunostaining scoring method and correlation with poor prognosis.

AIMS: c-Met is an emerging biomarker in pancreatic ductal adenocarcinoma (PDAC); there is no consensus regarding the immunostaining scoring method for this marker. We aimed to assess the prognostic value of c-Met overexpression in resected PDAC, and to elaborate a robust and reproducible scoring method for c-Met immunostaining in this setting. METHODS AND RESULTS: c-Met immunostaining was graded according to the validated MetMab score, a classic visual scale combining surface and intensity (SI score), or a simplified score (high c-Met: ≥20% of tumour cells with strong membranous staining), in stage I-II PDAC. A computer-assisted classification method (Aperio software) was developed. Clinicopathological parameters were correlated with disease-free survival (DFS) and overall survival(OS). One hundred and forty-nine patients were analysed retrospectively in a two-step process. Thirty-seven samples (whole slides) were analysed as a pre-run test. Reproducibility values were optimal with the simplified score (kappa = 0.773); high c-Met expression (7/37) was associated with shorter DFS [hazard ratio (HR) 3.456, P = 0.0036] and OS (HR 4.257, P = 0.0004). c-Met expression was concordant on whole slides and tissue microarrays in 87.9% of samples, and quantifiable with a specific computer-assisted algorithm. In the whole cohort (n = 131), patients with c-Met(high) tumours (36/131) had significantly shorter DFS (9.3 versus 20.0 months, HR 2.165, P = 0.0005) and OS (18.2 versus 35.0 months, HR 1.832, P = 0.0098) in univariate and multivariate analysis. CONCLUSIONS: Simplified c-Met expression is an independent prognostic marker in stage I-II PDAC that may help to identify patients with a high risk of tumour relapse and poor survival.