Agence Nationale De La Recherche Research Articles

Data-Driven Mergers

We study mergers between firms operating in data-connected markets: the data generated as a byproduct of the activity in market A can be used by firms operating in market B. The effects of such a merger depend on whether data trade among independent firms is possible and whether data use benefits consumers or leads to more surplus extraction. When data increases product B’s quality, the merger benefits consumers in both markets if data cannot be traded absent the merger and harms them otherwise. When data are used to extract consumer surplus in market B, the merger increases consumer surplus in market A and reduces it in market B. This paper was accepted by Joshua Gans, business strategy. Funding: This work was supported by Agence Nationale de la Recherche [Grant ANR-17-EURE-0010] (Investissements d’Avenir program) to A. de Cornière; Oxford Internet Institute’s Research Programme on AI, Government, and Policy, funded by the Dieter Schwarz Stiftung gGmbH to G. Taylor; and TSE Digital Center.

Are People Willing to Pay for Reduced Inequality?

Would consumers be willing to pay more for goods for which there is less inequality in wages across those involved in their production? In incentive-compatible behavioral choice studies on representative samples of the English and U.S. populations, we find significantly positive willingness to pay for such inequality reductions in more than 80% of subjects. Although it varies with political leaning and the extent of the inequality reduction, willingness to pay is positive across the political spectrum and for all studied inequality differences. It is higher for more intuitive and informative inequality-reporting formats. Our findings have policy implications for both governments and firms. On the one hand, they suggest the promise of universal provision of product-level inequality information as a tool for moderating income inequality. On the other hand, they highlight the potential relevance of inequality reporting for firms’ marketing strategies. This paper was accepted by Dorothea Kübler, behavioral economics and decision analysis. Funding: This work was supported by the Agence Nationale de la Recherche [LabEx Ecodec/ANR-11-LABX-0047] and the Fondation HEC [Budget F]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2023.01974 .

Marginal Values of a Stochastic Game

Zero-sum stochastic games are parameterized by payoffs, transitions, and possibly a discount rate. In this article, we study how the main solution concepts, the discounted and undiscounted values, vary when these parameters are perturbed. We focus on the marginal values, introduced by Mills in 1956 in the context of matrix games—that is, the directional derivatives of the value along any fixed perturbation. We provide a formula for the marginal values of a discounted stochastic game. Further, under mild assumptions on the perturbation, we provide a formula for their limit as the discount rate vanishes and for the marginal values of an undiscounted stochastic game. We also show, via an example, that the two latter differ in general. Funding: This work was supported by Fondation CFM pour la Recherche; the European Research Council [Grant ERC-CoG-863818 (ForM-SMArt)]; and Agence Nationale de la Recherche [Grant ANR-21-CE40-0020].

Multiomic analysis in fibroblasts of patients with inborn errors of cobalamin metabolism reveals concordance with clinical and metabolic variability

The high variability in clinical and metabolic presentations of inborn errors of cobalamin (cbl) metabolism (IECM), such as the cblC/epicblC types with combined deficits in methylmalonyl-coA mutase (MUT) and methionine synthase (MS), are not well understood. They could be explained by the impaired expression/activity of enzymes from other metabolic pathways. We performed metabolomic, genomic, proteomic, and post-translational modification (PTM) analyses in fibroblasts from three cblC cases and one epi-cblC case compared with three cblG cases with specific MS deficits and control fibroblasts. CblC patients had metabolic profilings consistent with altered urea cycle, glycine, and energy mitochondrial metabolism. Metabolomic analysis showed partial disruption and increased glutamate/ketoglutarate anaplerotic pathway of the tricarboxylic acid cycle (TCA), in patient fibroblasts. RNA-seq analysis showed decreased expression of MT-TT (mitochondrial tRNA threonine), MT-TP (mitochondrial tRNA proline), OXCT1 (succinyl CoA:3-oxoacid CoA transferase deficiency), and MT-CO1 (cytochrome C oxidase subunit 1). Proteomic changes were observed for key mitochondrial enzymes, including NADH:ubiquinone oxidoreductase subunit A8 (NDUFA8), carnitine palmitoyltransferase 2 (CPT2), and ubiquinol-cytochrome C reductase, complex III subunit X (UQCR10). Propionaldehyde addition in ornithine aminotransferase was the predominant PTM in cblC cells and could be related with the dramatic cellular increase in propionate and methylglyoxalate. It is consistent with the decreased concentration of ornithine reported in 3 cblC cases. Whether the changes detected after multi-omic analyses underlies clinical features in cblC and cblG types of IECM, such as peripheral and central neuropathy, cardiomyopathy, pulmonary hypertension, development delay, remains to be investigated. The omics-related effects of IECM on other enzymes and metabolic pathways are consistent with the diversity and variability of their age-related metabolic and clinical manifestations. PTMs are expected to produce cumulative effects, which could explain the influence of age on neurological manifestations. French Agence Nationale de la Recherche (Projects PREDICTS and EpiGONE) and Inserm.

The Iterates of the Frank–Wolfe Algorithm May Not Converge

The Frank–Wolfe algorithm is a popular method for minimizing a smooth convex function f over a compact convex set [Formula: see text]. Whereas many convergence results have been derived in terms of function values, almost nothing is known about the convergence behavior of the sequence of iterates [Formula: see text]. Under the usual assumptions, we design several counterexamples to the convergence of [Formula: see text], where f is d-time continuously differentiable, [Formula: see text], and [Formula: see text]. Our counterexamples cover the cases of open-loop, closed-loop, and line-search step-size strategies and work for any choice of the linear minimization oracle, thus demonstrating the fundamental pathologies in the convergence behavior of [Formula: see text]. Funding: The authors acknowledge the support of the AI Interdisciplinary Institute ANITI funding through the French “Investments for the Future – PIA3” program under the Agence Nationale de la Recherche (ANR) agreement [Grant ANR-19-PI3A0004], the Air Force Office of Scientific Research, Air Force Material Command, U.S. Air Force [Grants FA866-22-1-7012 and ANR MaSDOL 19-CE23-0017-0], ANR Chess [Grant ANR-17-EURE-0010], ANR Regulia, and Centre Lagrange.

Linear Program-Based Policies for Restless Bandits: Necessary and Sufficient Conditions for (Exponentially Fast) Asymptotic Optimality

We provide a framework to analyze control policies for the restless Markovian bandit model under both finite and infinite time horizons. We show that when the population of arms goes to infinity, the value of the optimal control policy converges to the solution of a linear program (LP). We provide necessary and sufficient conditions for a generic control policy to be (i) asymptotically optimal, (ii) asymptotically optimal with square root convergence rate, and (iii) asymptotically optimal with exponential rate. We then construct the LP-index policy that is asymptotically optimal with square root convergence rate on all models and with exponential rate if the model is nondegenerate in finite horizon and satisfies a uniform global attractor property in infinite horizon. We next define the LP-update policy, which is essentially a repeated LP-index policy that solves a new LP at each decision epoch. We conclude by providing numerical experiments to compare the efficiency of different LP-based policies. Funding: This work was supported by Agence Nationale de la Recherche [Grant ANR-19-CE23-0015].

An Experiment on Gender Representation in Majoritarian Bargaining

Women are underrepresented in business, academic, and political decision-making bodies across the world. To investigate the causal effect of gender representation on multilateral negotiations, we experimentally manipulate the composition of triads in a majoritarian, divide-the-dollar game. We document a robust gender gap in earnings driven largely by the exclusion of women from alliances rather than differential shares within alliances. Experiments with different subject pools show that distinct bargaining dynamics can underlie the same inequitable outcomes; gender-biased outcomes can be caused by outright discrimination, but they can also be driven by more complex dynamics related to differences in bargaining strategies. Although replacing the male with a female majority all but eliminates the gap in one pool, it has minimal effect in the other. These findings show that there is no “one-size-fits-all” solution to the gender gap we uncovered and highlight the importance of studying bargaining dynamics in detail. This paper was accepted by Marie-Claire Villeval, behavioral economics and decision analysis. Funding: A. Baranski recognizes financial support from Tamkeen [under the New York University Abu Dhabi Research Institute; Grant CG005]. J. Tremewan acknowledges funding from the Agence Nationale de la Recherche [Grant ANR-21-CE03-0018 (Project ENDURA)]. Supplemental Material: The online appendix and data files are available at https://doi.org/10.1287/mnsc.2022.01800 .

Mouse models of diabetes-related ulcers: a systematic review and network meta-analysis

Diabetic foot ulcers (DFUs) are a common complication of diabetes, associated with important morbidity. Appropriate animal models of DFUs may improve drug development, and subsequently the success rate of clinical trials. However, while many models have been proposed, they are extremely heterogeneous, and no standard has emerged. We thus propose a systematic review with a network meta-analysis (NMA) to gather direct and indirect evidence, and compare the different mouse models of diabetes-related ulcers. The systematic search was performed in Pubmed and Embase. The main outcomes were wound size measurement at days 3, 7, 11 and 15 (±1 day). The risk of bias and methodological quality of all included studies was assessed by using the Systematic Review Center for Laboratory animal Experimentation (SYRCLE) risk of bias tool. Meta-regressions were done on prespecified variables, including mouse strain, type of ulcer, sex, age, and use of a splint. We included 295 studies. Among all models, only db/db, ob/ob, streptozotocin (STZ), and STZ+high fat diet mice showed a significantly delayed wound healing, compared with controls, at each time point. Age, sex and ulcer type had influence on wound healing, although not at all time points. In conclusion, the db/db model is associated with the largest delay in wound healing The STZ model also exhibits significantly decreased wound healing. STZ+high fat diet and ob/ob mice may also be relevant models of diabetes-related ulcers, although the results rely on a more limited number of studies. This work was funded by the Agence Nationale de la Recherche (grant ANR-18-CE17-0017).

The burden of very early dropout in infertility care: a nationwide population-based cohort study.

What is the frequency and the associated factors of very early dropout following unsuccessful clomiphene citrate (CC)/gonadotropin treatment in the context of full coverage of treatment cost. Despite free treatment, almost one in four women had a very early dropout following unsuccessful CC/gonadotropin treatment, with patients below the poverty line being more likely to drop out early. Success of infertility care is tarnished by very high dropout rates. Infertility care dropout has been considered as resulting principally from financial barriers because of the high cost of treatment. Nearly all previous work addressed dropout following IVF/ICSI. Factors associated with dropout following CC/gonadotropins may be different and also need to be investigated. Nationwide population-based cohort study. Using the French national health insurance and hospital databases, we included in the cohort 27 416 women aged 18-49 years unsuccessfully treated with CC/gonadotropins in 2017. The main outcome was very early dropout, defined as discontinuation of all infertility treatment after unsuccessful treatment for 1-3 months. Very early treatment dropout was analysed by multivariate logistic regression. Among women unsuccessfully treated with CC/gonadotropins, 22% dropped out of infertility care within 3 months. In multivariate analysis, higher early dropout following unsuccessful CC/gonadotropin treatment was associated with older and younger ages (≥35 and <25 years), being below the poverty line, being treated with CC prescribed by a general practitioner and lack of infertility tests or monitoring. This study is based on health administrative data that do not include reasons for dropout and record only a limited amount of information. It is thus not possible to analyse the reason for early dropout. Despite full coverage of all infertility treatment, women under the poverty line have a higher risk of very early dropout following unsuccessful CC/gonadotropin treatment. Better understanding is needed of the non-financial barriers and difficulties faced by these patients. To address disparities in infertility treatment, practitioner training could be reinforced to adapt to patients from different social and cultural backgrounds. This work was supported by the ANR StimHo project, grant ANR-17-CE36-0011-01 from the French Agence Nationale de la Recherche. The authors have no conflict of interest to declare. N/A.

Impact of piperaquine resistance in Plasmodium falciparum on malaria treatment effectiveness in The Guianas: a descriptive epidemiological study

Plasmodium falciparum is an apicomplexan parasite responsible for lethal cases of malaria. According to WHO recommendations, P falciparum cases are treated with artemisinin-based combination therapy including dihydroartemisinin-piperaquine. However, the emergence of resistant parasites against dihydroartemisinin-piperaquine was reported in southeast Asia in 2008 and, a few years later, suspected in South America. To characterise resistance emergence, a treatment efficacy study was performed on the reported patients infected with P falciparum and treated with dihydroartemisinin-piperaquine in French Guiana (n=6, 2016-18). Contemporary isolates collected in French Guiana were genotyped for P falciparum chloroquine resistance transporter (pfCRT; n=845) and pfpm2 and pfpm3 copy number (n=231), phenotyped using the in vitro piperaquine survival assay (n=86), and analysed through genomic studies (n=50). Additional samples from five Amazonian countries and one outside the region were genotyped (n=1440). In field isolates, 40 (47%) of 86 (95% CI 35·9-57·1) were resistant to piperaquine in vitro; these phenotypes were more associated with pfCRTC350R (ie, Cys350Arg) and pfpm2 and pfpm3 amplifications (Dunn test, p<0·001). Those markers were also associated with dihydroartemisinin-piperaquine treatment failure (n=3 [50%] of 6). A high prevalence of piperaquine resistance markers was observed in Suriname in 19 (83%) of 35 isolates and in Guyana in 579 (73%) of 791 isolates. The pfCRTC350R mutation emerged before pfpm2 and pfpm3 amplification in a temporal sequence different from southeast Asia, and in the absence of artemisinin partial resistance, suggesting a geographically distinctive epistatic relationship between these genetic markers. The high prevalence of piperaquine resistance markers in parasite populations of the Guianas, and the risk of associated therapeutic failures calls for caution on dihydroartemisinin-piperaquine use in the region. Furthermore, greater attention should be given to potential differences in genotype to phenotype mapping across genetically distinct parasite populations from different continents. Pan American Health Organization and WHO, French Ministry for Research, European Commission, Santé publique France, Agence Nationale de la Recherche, Fundação de Amparo à Pesquisa do Estado do Amazonas, Ministry of Health of Brazil, Oswaldo Cruz Foundation, and National Institutes of Health. For the French and Portuguese translations of the abstract see Supplementary Materials section.

Privacy-Preserving Personalized Revenue Management

This paper examines how data-driven personalized decisions can be made while preserving consumer privacy. Our setting is one in which the firm chooses a personalized price based on each new customer’s vector of individual features; the true set of individual demand-generating parameters is unknown to the firm and so must be estimated from historical data. We extend the existing personalized pricing framework by requiring also that the firm’s pricing policy preserve consumer privacy, or (formally) that it be differentially private: an industry standard for privacy preservation. We develop privacy-preserving personalized pricing algorithms and show that they achieve near-optimal revenue by deriving theoretical (upper and lower) performance bounds. Our analyses further suggest that, if the firm possesses a sufficient amount of historical data, then it can achieve a certain level of differential privacy almost “for free.” That is, the revenue loss due to privacy preservation is of smaller order than that due to estimation. We confirm our theoretical findings in a series of numerical experiments based on synthetically generated and online auto lending (CPRM-12-001) data sets. Finally, motivated by practical considerations, we also extend our algorithms and findings to a variety of alternative settings, including multiproduct pricing with substitution effect, discrete feasible price set, categorical sensitive features, and personalized assortment optimization. This paper was accepted by Vishal Gaur, operations management. Funding: R. Momot acknowledges financial support from the HEC Paris Foundation and the Agence Nationale de la Recherche (French National Research Agency) “Investissements d’Avenir” [Grant LabEx Ecodec/ANR-11-LABX-0047] during the initial stages of this project. Supplemental Material: The data files and online appendices are available at https://doi.org/10.1287/mnsc.2023.4925 .

Long-COVID cognitive impairments and reproductive hormone deficits in men may stem from GnRH neuronal death

We have recently demonstrated a causal link between loss of gonadotropin-releasing hormone (GnRH), the master molecule regulating reproduction, and cognitive deficits during pathological aging, including Down syndrome and Alzheimer's disease. Olfactory and cognitive alterations, which persist in some COVID-19 patients, and long-term hypotestosteronaemia in SARS-CoV-2-infected men are also reminiscent of the consequences of deficient GnRH, suggesting that GnRH system neuroinvasion could underlie certain post-COVID symptoms and thus lead to accelerated or exacerbated cognitive decline. We explored the hormonal profile of COVID-19 patients and targets of SARS-CoV-2 infection in post-mortem patient brains and human fetal tissue. We found that persistent hypotestosteronaemia in some men could indeed be of hypothalamic origin, favouring post-COVID cognitive or neurological symptoms, and that changes in testosterone levels and body weight over time were inversely correlated. Infection of olfactory sensory neurons and multifunctional hypothalamic glia called tanycytes highlighted at least two viable neuroinvasion routes. Furthermore, GnRH neurons themselves were dying in all patient brains studied, dramatically reducing GnRH expression. Human fetal olfactory and vomeronasal epithelia, from which GnRH neurons arise, and fetal GnRH neurons also appeared susceptible to infection. Putative GnRH neuron and tanycyte dysfunction following SARS-CoV-2 neuroinvasion could be responsible for serious reproductive, metabolic, and mental health consequences in long-COVID and lead to an increased risk of neurodevelopmental and neurodegenerative pathologies over time in all age groups. European Research Council (ERC) grant agreements No 810331, No 725149, No 804236, the European Union Horizon 2020 research and innovation program No 847941, the Fondation pour la Recherche Médicale (FRM) and the Agence Nationale de la Recherche en Santé (ANRS) No ECTZ200878 Long Covid 2021 ANRS0167 SIGNAL, Agence Nationale de la recherche (ANR) grant agreements No ANR-19-CE16-0021-02, No ANR-11-LABEX-0009, No. ANR-10-LABEX-0046, No. ANR-16-IDEX-0004, Inserm Cross-Cutting Scientific Program HuDeCA, the CHU Lille Bonus H, the UK Medical Research Council (MRC) and National Institute of Health and care Research (NIHR).

O.5.1-3 Co-creation of an active transportation promotion intervention: a concept mapping study

Abstract Purpose Active transportation (AT) is an efficient way to increase daily physical activity and reduce pollution from the transportation sector. Mhealth has contributed to behavior change in fields like physical activity and has been explored to encourage AT. Most interventions, however, did not incorporate evidence-based behavior change techniques and were not co-created with potential users. We aim to understand how AT can be promoted via mhealth intervention by using a bottom-up participatory approach. Methods Concept mapping was done online via GroupWisdom platform, with the public of Alpes-Maritimes County, France. Participants were asked to brainstorm, sort, and rate ideas about mobile app features to encourage AT. Multidimensional scaling algorithm and hierarchical cluster analysis were used to create visual cluster maps to illustrate the participants’ conceptual thinking of the topic. We compared the data by participants’ age, education, and typical transportation mode to understand how different groups perceived each idea and clusters. Results Participants generated 44 unique ideas that they believe would encourage AT behaviors. Six themes emerged, from highest to lowest cluster rating: Infrastructure (n = 9 ideas) and Itinerary (n = 9 ideas), which were consistently rated across groups; Contact with government (n = 4 ideas); Data (n = 12 ideas), with the most number of ideas generated; App usability (n = 6 ideas); and Legislation and code of conduct (n = 4 ideas), with the least number of ideas. Conclusions Our findings showed various ways AT can be promoted. Even though the participants were asked about mobile apps features, they quickly pointed out ideas unrelated to mobile apps. The built environment and infrastructure are clearly the most important factors to promote AT, for our participants. However, they have also identified other soft measures like information, participation in decision making, and social aspects that can contribute to behavior change and support sustainable and healthy modes of transportation. Therefore, a systemic and multifaceted approach is recommended to promote AT. Support/Funding Source This work was supported by the French government, managed by the Agence Nationale de la Recherche under the project Investissements d'Avenir UCAJEDI with the reference n° ANR-15-IDEX-01.

Prediction of cesarean delivery in class III obese nulliparous women: An externally validated model using machine learning.

class III obese women, are at a higher risk of cesarean section during labor, and cesarean section is responsible for increased maternal and neonatal morbidity in this population. the objective of this project was to develop a method with which to quantify cesarean section risk before labor. this is a multicentric retrospective cohort study conducted on 410 nulliparous class III obese pregnant women who attempted vaginal delivery in two French university hospitals. We developed two predictive algorithms (a logistic regression and a random forest models) and assessed performance levels and compared them. the logistic regression model found that only initial weight and labor induction were significant in the prediction of unplanned cesarean section. The probability forest was able to predict cesarean section probability using only two pre-labor characteristics: initial weight and labor induction. Its performances were higher and were calculated for a cut-point of 49.5% risk and the results were (with 95% confidence intervals): area under the curve 0.70 (0.62,0.78), accuracy 0.66 (0.58, 0.73), specificity 0.87 (0.77, 0.93), and sensitivity 0.44 (0.32, 0.55). this is an innovative and effective approach to predicting unplanned CS risk in this population and could play a role in the choice of a trial of labor versus planned cesarean section. Further studies are needed, especially a prospective clinical trial. French state funds "Plan Investissements d'Avenir" and Agence Nationale de la Recherche.

Development and validation of an interpretable machine learning-based calculator for predicting 5-year weight trajectories after bariatric surgery: a multinational retrospective cohort SOPHIA study

Weight loss trajectories after bariatric surgery vary widely between individuals, and predicting weight loss before the operation remains challenging. We aimed to develop a model using machine learning to provide individual preoperative prediction of 5-year weight loss trajectories after surgery. In this multinational retrospective observational study we enrolled adult participants (aged ≥18 years) from ten prospective cohorts (including ABOS [NCT01129297], BAREVAL [NCT02310178], the Swedish Obese Subjects study, and a large cohort from the Dutch Obesity Clinic [Nederlandse Obesitas Kliniek]) and two randomised trials (SleevePass [NCT00793143] and SM-BOSS [NCT00356213]) in Europe, the Americas, and Asia, with a 5 year follow-up after Roux-en-Y gastric bypass, sleeve gastrectomy, or gastric band. Patients with a previous history of bariatric surgery or large delays between scheduled and actual visits were excluded. The training cohort comprised patients from two centres in France (ABOS and BAREVAL). The primary outcome was BMI at 5 years. A model was developed using least absolute shrinkage and selection operator to select variables and the classification and regression trees algorithm to build interpretable regression trees. The performances of the model were assessed through the median absolute deviation (MAD) and root mean squared error (RMSE) of BMI. 10 231 patients from 12 centres in ten countries were included in the analysis, corresponding to 30 602 patient-years. Among participants in all 12 cohorts, 7701 (75·3%) were female, 2530 (24·7%) were male. Among 434 baseline attributes available in the training cohort, seven variables were selected: height, weight, intervention type, age, diabetes status, diabetes duration, and smoking status. At 5 years, across external testing cohorts the overall mean MAD BMI was 2·8 kg/m2 (95% CI 2·6-3·0) and mean RMSE BMI was 4·7 kg/m2 (4·4-5·0), and the mean difference between predicted and observed BMI was -0·3 kg/m2 (SD 4·7). This model is incorporated in an easy to use and interpretable web-based prediction tool to help inform clinical decision before surgery. We developed a machine learning-based model, which is internationally validated, for predicting individual 5-year weight loss trajectories after three common bariatric interventions. SOPHIA Innovative Medicines Initiative 2 Joint Undertaking, supported by the EU's Horizon 2020 research and innovation programme, the European Federation of Pharmaceutical Industries and Associations, Type 1 Diabetes Exchange, and the Juvenile Diabetes Research Foundation and Obesity Action Coalition; Métropole Européenne de Lille; Agence Nationale de la Recherche; Institut national de recherche en sciences et technologies du numérique through the Artificial Intelligence chair Apprenf; Université de Lille Nord Europe's I-SITE EXPAND as part of the Bandits For Health project; Laboratoire d'excellence European Genomic Institute for Diabetes; Soutien aux Travaux Interdisciplinaires, Multi-établissements et Exploratoires programme by Conseil Régional Hauts-de-France (volet partenarial phase 2, project PERSO-SURG).

A New Family of Route Formulations for Split Delivery Vehicle Routing Problems

We propose a new family of formulations with route-based variables for the split delivery vehicle routing problem with and without time windows. Each formulation in this family is characterized by the maximum number of different demand quantities that can be delivered to a customer during a vehicle visit. As opposed to previous formulations in the literature, the exact delivery quantities are not always explicitly known in this new family. The validity of these formulations is ensured by an exponential set of nonrobust constraints. Additionally, we explore a property of optimal solutions that enables us to determine a minimum delivery quantity based on customer demand and vehicle capacity, and this number is often greater than one. We use this property to reduce the number of possible delivery quantities in our formulations, improving the solution times of the computationally strongest formulation in the family. Furthermore, we propose new variants of nonrobust cutting planes that strengthen the formulations, namely limited-memory subset-row covering inequalities and limited-memory strong k-path inequalities. Finally, we develop a branch-cut-and-price (BCP) algorithm to solve our formulations enriched with the proposed valid inequalities, which resorts to state-of-the-art algorithmic enhancements. We show how to effectively manage the nonrobust cuts when solving the pricing problem that dynamically generates route variables. Numerical results indicate that our formulations and BCP algorithm establish new state-of-the-art results for the variant with time windows, as many benchmark instances with 50 and 100 customers are solved to optimality for the first time. Several instances of the variant without time windows are solved to proven optimality for the first time. Funding: This work was supported by the Conselho Nacional de Desenvolvimento Científico e Tecnológico [Grants 306033/2019-4, 313220/2020-4, and 314088/2021-0], the Région Nouvelle Aquitaine, France [Grant AAPR2020A-2020-8601810], the Agence Nationale de la Recherche [Grant ANR-20-CE40-0021-01], the Fundação de Amparo à Pesquisa do Estado de São Paulo [Grants 13/07375-0, 16/01860-1, and 19/23596-2], and the Paraíba State Research Foundation [Grants 261/2020 and 041/2023]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/trsc.2022.0085 .

Tryptophan metabolism promotes immune evasion in human pancreatic β cells

To resist the autoimmune attack characteristic of type 1 diabetes, insulin producing pancreatic β cells need to evade T-cell recognition. Such escape mechanisms may be conferred by low HLA class I (HLA-I) expression and upregulation of immune inhibitory molecules such as Programmed cell Death Ligand 1 (PD-L1). The expression of PD-L1, HLA-I and CXCL10 was evaluated in the human β cell line, ECN90, and in primary human and mouse pancreatic islets. Most genes were determined by real-time RT-PCR, flow cytometry and Western blot. Activator and inhibitor of the AKT signaling were used to modulate PD-L1 induction. Key results were validated by monitoring activity of CD8+ Jurkat T cells presenting β cell specific T-cell receptor and transduced with reporter genes in contact culture with the human β cell line, ECN90. In this study, we identify tryptophan (TRP) as an agonist of PD-L1 induction through the AKT signaling pathway. TRP also synergistically enhanced PD-L1 expression on β cells exposed to interferon-γ. Conversely, interferon-γ-mediated induction of HLA-I and CXCL10 genes was down-regulated upon TRP treatment. Finally, TRP and its derivatives inhibited the activation of islet-reactive CD8+ T cells by β cells. Collectively, our findings indicate that TRP could induce immune tolerance to β cells by promoting their immune evasion through HLA-I downregulation and PD-L1 upregulation. Dutch Diabetes Research Foundation, DON Foundation, the Laboratoire d'Excellence consortium Revive (ANR-10-LABX-0073), Agence Nationale de la Recherche (ANR-19-CE15-0014-01), Fondation pour la Recherche Médicale (EQ U201903007793-EQU20193007831), Innovative Medicines InitiativeINNODIA and INNODIA HARVEST, Aides aux Jeunes Diabetiques (AJD) and Juvenile Diabetes Research Foundation Ltd (JDRF).

Combining robust urine biomarkers to assess chronic kidney disease progression.

Urinary biomarkers may improve the prediction of chronic kidney disease (CKD) progression. Yet, data reporting the applicability of most commercial biomarker assays to the detection of their target analyte in urine together with an evaluation of their predictive performance are scarce. 30 commercial assays (ELISA) were tested for their ability to quantify the target analyte in urine using strict (FDA-approved) validation criteria. In an exploratory analysis, LASSO (Least Absolute Shrinkage and Selection Operator) logistic regression analysis was used to identify potentially complementary biomarkers predicting fast CKD progression, determined as the 51CrEDTA clearance-based measured glomerular filtration rate (mGFR) decline (>10% per year) in a subsample of 229 CKD patients (mean age, 61 years; 66% men; baseline mGFR, 38mL/min) from the NephroTest prospective cohort. Among the 30 assays, directed against 24 candidate biomarkers, encompassing different pathophysiological mechanisms of CKD progression, 16 assays fulfilled the FDA-approved criteria. LASSO logistic regressions identified a combination of five biomarkers including CCL2, EGF, KIM1, NGAL, and TGF-α that improved the prediction of fast mGFR decline compared to the kidney failure risk equation variables alone: age, gender, mGFR, and albuminuria. Mean area under the curves (AUC) estimated from 100 re-samples was higher in the model with than without these biomarkers, 0.722 (95% confidence interval 0.652-0.795) vs. 0.682 (0.614-0.748), respectively. Fully-adjusted odds-ratios (95% confidence interval) for fast progression were 1.87 (1.22, 2.98), 1.86 (1.23, 2.89), 0.43 (0.25, 0.70), 1.10 (0.71, 1.83), 0.55 (0.33, 0.89), and 2.99 (1.89, 5.01) for albumin, CCL2, EGF, KIM1, NGAL, and TGF-α, respectively. This study provides a rigorous validation of multiple assays for relevant urinary biomarkers of CKD progression which combination may improve the prediction of CKD progression. This work was supported by Institut National de la Santé et de la Recherche Médicale, Université de Paris, Assistance Publique Hôpitaux de Paris, Agence Nationale de la Recherche, MSDAVENIR, Pharma Research and Early Development Roche Laboratories (Basel, Switzerland), and Institut Roche de Recherche et Médecine Translationnelle (Paris, France).

Solving Optimization Problems with Blackwell Approachability

In this paper, we propose a new algorithm for solving convex-concave saddle-point problems using regret minimization in the repeated game framework. To do so, we introduce the Conic Blackwell Algorithm+ ([Formula: see text]), a new parameter- and scale-free regret minimizer for general convex compact sets. [Formula: see text] is based on Blackwell approachability and attains [Formula: see text] regret. We show how to efficiently instantiate [Formula: see text] for many decision sets of interest, including the simplex, [Formula: see text] norm balls, and ellipsoidal confidence regions in the simplex. Based on [Formula: see text], we introduce [Formula: see text], a new parameter-free algorithm for solving convex-concave saddle-point problems achieving a [Formula: see text] ergodic convergence rate. In our simulations, we demonstrate the wide applicability of [Formula: see text] on several standard saddle-point problems from the optimization and operations research literature, including matrix games, extensive-form games, distributionally robust logistic regression, and Markov decision processes. In each setting, [Formula: see text] achieves state-of-the-art numerical performance and outperforms classical methods, without the need for any choice of step sizes or other algorithmic parameters. Funding: J. Grand-Clément is supported by the Agence Nationale de la Recherche [Grant 11-LABX-0047] and by Hi! Paris. C. Kroer is supported by the Office of Naval Research [Grant N00014-22-1-2530] and by the National Science Foundation [Grant IIS-2147361].

Large Ranking Games with Diffusion Control

We consider a symmetric stochastic differential game where each player can control the diffusion intensity of an individual dynamic state process, and the players whose states at a deterministic finite time horizon are among the best [Formula: see text] of all states receive a fixed prize. Within the mean field limit version of the game, we compute an explicit equilibrium, a threshold strategy that consists of choosing the maximal fluctuation intensity when the state is below a given threshold and the minimal intensity otherwise. We show that for large n, the symmetric n-tuple of the threshold strategy provides an approximate Nash equilibrium of the n-player game. We also derive the rate at which the approximate equilibrium reward and the best-response reward converge to each other, as the number of players n tends to infinity. Finally, we compare the approximate equilibrium for large games with the equilibrium of the two-player case. Funding: Support from the Deutsche Forschungsgemeinschaft [Grant AN 1024/5-1] is acknowledged. The research of N. Kazi-Tani is supported by the Agence Nationale de la Recherche [Grant ANR-21-CE46-0002-03]. The research of C. Zhou is supported by the National Natural Science Foundation of China [Grant 11871364], the Singapore Ministry of Education [Grants A-8000453-00-00, A-0004273-00-00 and A-0004277-00-00] and the MERLION 2020 award [Grant A-0004589-00-00].