2025Activity‌​‌ reportProject-TeamAISTROSIGHT

2.1 Failures in‌​‌ drug development for neurological​​ diseases

The set of​​​‌ available drugs for neurological‌ diseases is both aging‌​‌ and lacking in effectiveness.​​ There remains a very​​​‌ high unmet medical need‌ for treatments in neurology,‌​‌ despite heavy historical investment​​ in the field 29​​​‌, 60. A‌ typical drug development process‌​‌ includes a set of​​ successive stages (fig. 1​​​‌), sequentially evaluating the‌ effects of the candidate‌​‌ drug in vitro on​​ cell culture models, then​​​‌ in vivo on animal‌ models (pre-clinical), after which‌​‌ the mechanistic origins of​​ the candidate drug effect​​​‌ can be assessed in‌ animals (sometimes in humans).‌​‌ The next stage consists​​ of studies in humans​​​‌ based on clinical trials‌ that are themselves structured‌​‌ in successive phases: phase​​ I to test for​​​‌ safety, phase II to‌ determine the effect of‌​‌ the candidate on a​​ small set of patients​​​‌ and phase III that‌ includes large cohorts of‌​‌ patients and control people​​ in a randomized setting.​​​‌ Each and every stage‌ of this process has‌​‌ significant probability to fail​​ and break off the​​​‌ development of the drug‌ under process. With the‌​‌ recent public health issues​​ (HIV, Covid-19) the general​​​‌ public has mostly been‌ made aware of failures‌​‌ between the successive phases​​ of the clinical trial​​​‌ stage. However, in the‌ field of neurology, the‌​‌ difficulties in developing drug​​ candidates are mainly due​​​‌ to a high failure‌ rate in the clinic:‌​‌ the activity of the​​ drug candidate in in​​​‌ vitro cell cultures or‌ in animal models is‌​‌ very often not confirmed​​ in humans 44,​​​‌ 48.

In recent‌ years, numerical approaches have‌​‌ been proposed in the​​ field, either with mechanistic​​​‌ modeling to predict the‌ response of the cell‌​‌ to the candidate molecule​​ (quantitative systems biology/pharmacology) 55​​​‌, 61 or with‌ machine learning to identify‌​‌ the impacted (sub)cellular systems​​ or the effects of​​​‌ the candidate drug 69‌, 50. However,‌​‌ these approaches are still​​ inefficient to meet the​​​‌ above challenges because they‌ often address a unique‌​‌ scale or modality of​​ interest (e.g., molecular, cellular,​​​‌ preclinical) and lose their‌ predictive power at other‌​‌ scales (e.g. clinical, i.e.​​ the patient). The main​​​‌ methodological objective of AIstroSight‌ is to develop quantitative‌​‌ systems biology and Artificial​​ Intelligence (AI) approaches able​​​‌ to embrace several of‌ these scales.

A schematic‌​‌ view of a drug​​ development process

2.2 Main deliverables​​

Our overall goal is​​​‌ to develop innovative numerical‌ methods for neuropharmacology that‌​‌ will provide us with​​ levers to accelerate and​​​‌ derisk the early stages‌ of drug design. As‌​‌ a main deliverable and​​ proof of concept of​​​‌ the efficiency of these‌ methods, our ambition for‌​‌ the first four years​​ of the project is​​​‌ to identify a handful‌ (2 to 4) of‌​‌ new candidate drugs against​​​‌ neurological diseases.

2.3 Overview​ of the AIstroSight roadmap​‌

To improve the probability​​ of success of drug​​​‌ candidates in neurology, we​ integrate complementary information offered​‌ by data harvested at​​ different spatio-temporal scales (fig.​​​‌ 2): from the​ inside of the cell​‌ (molecular and cellular biology)​​ to the whole brain​​​‌ (imaging) and even to​ a population of patients​‌ (hospital data), using numerical​​ tools coupling mechanistic models​​​‌ with dedicated AI approaches.​ In a way, our​‌ strategy is to break​​ down the classical stratification​​​‌ silo of Fig. 1​, in which literature​‌ search, in vitro cell​​ culture, in vivo preclinic​​​‌ studies and in vivo​ clinic studies are viewed​‌ as a sequential multi-stage​​ process. Instead, we propose​​​‌ an integrated machine learning​ framework into which all​‌ those data are combined​​ to predict the effect​​​‌ of a candidate drug​ molecule.

AIstroSight develops innovative​‌ numerical approaches to integrate​​ these information sources into​​​‌ a coherent stream of​ data and expert knowledge,​‌ combining the analysis of​​ experimental observations with reasoning​​​‌ (of different kinds). Currently,​ these tasks are carried​‌ out in isolation and​​ their reconciliation is devolved​​​‌ to biologists/physicians. The originality​ of the AIstroSight contributions​‌ are approaches that automatically​​ carry out this reconciliation​​​‌ to assist biologists/physicians.

A​ schematic for the scientific​‌ roadmap of AIstroSight

Since AI​‌ algorithms are often black-box​​ tools, we also develop​​​‌ mechanistic modeling approaches (multiscale​ quantitative systems biology/pharmacology) to​‌ produce explanations for the​​ predictions of the AI​​​‌ algorithms, that can be​ rooted in neurobiology. Another​‌ important aspect of AIstroSight​​ is to widen the​​​‌ focus of neuropharmacology beyond​ neurons, that constitute only​‌ one half of the​​ nerve cells in the​​​‌ brain, and also take​ into account the other​‌ half, that is made​​ up by glial cells​​​‌ and their interactions with​ neurons. In particular, we​‌ consider the pharmacology of​​ astrocytes 71, one​​​‌ major subtype of glial​ cells, in interaction with​‌ the pharmacology of neurons.​​

2.4 Principles of the​​​‌ AIstroSight partnership

To accelerate​ cross-fertilization between digital science​‌ and medical research, AIstroSight​​ will be located on​​​‌ the East Hospital Campus​ of the Lyon University​‌ Hospital, the “Hospices Civils​​ de Lyon” (HCL), from​​​‌ 2024. We will also​ benefit from our strong​‌ association with CERMEP,​​ the preclinical and clinical​​​‌ in vivo imaging platform​ of the HCL. In​‌ 2024, the whole team​​ is indeed expected to​​​‌ move to Lyon's neurology​ hospital, located just across​‌ tA joint team with​​ the HCLhe street of​​​‌ CERMEP.

CERMEP is also​ affiliated with University Claude​‌ Bernard Lyon 1, Inserm​​ and CNRS. This provides​​​‌ us with an exceptional​ environment for the engineering​‌ of brain biochemical imaging​​ methods that allow the​​​‌ study of the effect​ of molecules on the​‌ whole brain (fMRI, PET,​​ fUS) and the analysis​​​‌ methodologies for these images.​ The CERMEP also hosts​‌ team BIORAN of the​​ “Centre de Recherche en​​​‌ Neurosciences de Lyon” (CRNL)​ laboratory, that has expertise​‌ ranging from the chemistry​​ of candidate-molecules to their​​ biochemical assays, from radiolabelling​​​‌ to animal PET/MRI imaging‌ and from preclinical models‌​‌ to first-in-man studies in​​ patients. The modeling expertise​​​‌ on the binding between‌ candidate molecules and receptors‌​‌ (structural biology, docking) is​​ also present at the​​​‌ CERMEP.

As a joint‌ team with the HCL,‌​‌ part of AIstroSight technology​​ developments is intended to​​​‌ be integrated into the‌ hospital information system developed‌​‌ during the last decade​​ by the HCL for​​​‌ patient management. This is‌ in particular the case‌​‌ of the development of​​ “Multi-patient query for care​​​‌ pathway characterization and clinical‌ trials”. Beyond participating in‌​‌ the HCL's mission as​​ an innovation leader in​​​‌ digital health, AIstroSight also‌ represents an opportunity for‌​‌ the HCL to reinforce​​ its infrastructure for the​​​‌ organization of clinical trials,‌ for instance in cooperation‌​‌ with pharma/biotech companies like​​ Theranexus. Like other teams​​​‌ of Inria Lyon, AIstroSight‌ is intensively implicated in‌​‌ the “AI innovation department”​​ (“Pôle de Développement IA”)​​​‌ that Inria Lyon and‌ HCL are supporting.

Finally,‌​‌ to ensure the impact​​ of our works on​​​‌ pharmacology and provide it‌ with potential industrial exit‌​‌ routes, the AIstroSight partnership​​ also includes an industrial​​​‌ partner, Theranexus, a‌ French biopharmaceutical company (an‌​‌ SME) that develops drugs​​ for the treatment of​​​‌ nervous system diseases with‌ an original focus on‌​‌ both neurons and astrocytes.​​ Theranexus is listed on​​​‌ the Euronext Growth market‌ in Paris and its‌​‌ social headquarters are located​​ in Lyon. A biotech​​​‌ company, the expertise of‌ its members is entirely‌​‌ in the experimental aspects,​​ not in the digital.​​​‌ Theranexus brings to AIstroSight‌ experimental data (experimental cell‌​‌ biology and brain imaging​​ for pharmacology), and provides​​​‌ their expertise for the‌ development of the digital‌​‌ tools needed to analyze​​ these data. In return,​​​‌ the objective is that‌ the output of these‌​‌ digital tools reveal novel​​ drug targets or novel​​​‌ candidate molecules that Theranexus‌ may decide to use‌​‌ to develop new treatments,​​ starting with the necessary​​​‌ clinical trials. Importantly, the‌ fact that these candidate‌​‌ drugs have been selected​​ from an innovative numerical​​​‌ approach strongly consolidates the‌ credibility of their development‌​‌ on the pharmaceutics market.​​ In addition, Theranexus brings​​​‌ to AIstroSight its know-how‌ and industrial expertise on‌​‌ the development of drug​​ candidates up to the​​​‌ market and its strategic‌ knowledge of the neuropharmacology‌​‌ industry. In this operating​​ scheme, Theranexus is therefore​​​‌ the preferred partner for‌ the early-phase transfer of‌​‌ the molecules that AIstroSight​​ could identify.

Independenlty from​​​‌ AIstroSight, Theranexus and BIORAN‌ have a longstanding collaboration‌​‌ together, in particular in​​ the framework of an​​​‌ ANR- and AURA Region-funded‌ joint laboratory (LabCom) called‌​‌ « NeuroImaging for Drug​​ Discovery (NI2D)» that aims​​​‌ at the development of‌ gliopharmacology using preclinical imaging‌​‌ tools (PET/MRI/brain ultrasound). This​​ LabCom is also hosted​​​‌ within the CERMEP premises.‌ NI2D aim at developing‌​‌ preclinical neuroimaging techniques (in​​ animals, mainly fMRI, PET​​​‌ and fUS = Functional‌ Ultra-Sound) for the evaluation‌​‌ of drug-candidates. AIstroSight develops​​ numerical methods capable of​​​‌ integrating data at multiple‌ scales for pharmacology, data‌​‌ that include imaging but​​​‌ also molecular data (intracellular​ signaling, omics data) or​‌ clinical data (biology, treatments,​​ medico-administrative). We thus benefits​​​‌ from the imaging data​ and methodologies of NI2D.​‌ The two are therefore​​ complementary, especially since we​​​‌ both share a strong​ interest in neuron/astrocyte interactions.​‌

3.1​​ Characterizing the mechanisms of​​​‌ action of candidate drugs​

Drug screening, either in​‌ vitro or in silico,​​ generally does not provide​​​‌ an explanation of the​ mechanism by which the​‌ identified drug acts at​​ the cellular level. However,​​​‌ this information is crucial​ (e.g., with respect to​‌ patients or health agencies),​​ and the algorithms used​​​‌ for the screening must​ be made explicable. Our​‌ goal here is to​​ use mathematical models and​​​‌ their hybridization with machine​ learning to provide explanations​‌ on the mechanisms of​​ action of a candidate​​​‌ molecule.

We develop mechanistic​ models of regulatory networks​‌ or intracellular signaling pathways​​ specific to the action​​​‌ of the candidate drugs​ identified by the screening.​‌ Those models predict the​​ spatio-temporal evolution of the​​​‌ concentrations of the molecular​ species involved in the​‌ modelled pathways using classical​​ reaction terms from biochemical​​​‌ kinetics and mass-action laws​ (first order reactions, bi-molecular​‌ reactions, Michaelis-Menten, Hill kinetics…).​​ Depending on the importance​​​‌ of intracellular spatial gradients​ and biochemical noise, space​‌ and stochasticity is accounted​​ for, thus resulting in​​​‌ models based on reaction-diffusion​ equations, stochastic or ordinary​‌ differential equations, or other​​ related formalisms (Gillespie algorithm,​​​‌ flow analysis). These models​ allow us to simulate​‌ in time and/or space​​ of the cell the​​​‌ mechanisms that govern the​ dynamics of the implicated​‌ molecules and how this​​ dynamic is altered by​​​‌ the selected drug. The​ aim is to use​‌ such mechanistic modeling to​​ produce explanations for the​​​‌ predictions made by the​ statistical learning techniques that​‌ are used in the​​ other sections. It is​​​‌ unlikely that these mechanistic​ models in themselves allow​‌ us to decipher the​​ totality of the molecular​​​‌ mechanisms involved, but they​ provide critical information to​‌ properly adjust the laboratory​​ and clinical experiments.

To​​​‌ be efficient, this approach​ demands that we maintain​‌ an effective knowledge basis​​ and expertise on the​​​‌ fundamental molecular mechanisms at​ play at these spatial​‌ scales and their modelling.​​ To build and maintain​​​‌ this expertise, we rely​ on existing long-term collaborations​‌ between AIstroSight members and​​ experimental neuroscientists -electrophysiologists- or​​​‌ neuropharmacologists on the intracellular​ signaling networks at play​‌ in neuron function or​​ neuro-astrocyte interactions.

- Agonist​​​‌ bias in GPCR:​ G protein-coupled receptors (GPCR)​‌ are currently the largest​​ family of molecular targets​​​‌ for potential new drugs​ 26. GPCR are​‌ cell-membrane receptors ubiquitously found​​ in all mammalian cells,​​​‌ but in particular in​ brain cells (neurons and​‌ astrocytes), where they control​​ a large repertoire of​​​‌ neuronal and astrocytic responses​ to a variety of​‌ external stimuli and molecules.​​ Bias antagonism refers to​​​‌ the observation that two​ ligands of the same​‌ GPCR can activate very​​ different cell responses 51​​​‌. This phenomenon is​ still not understood, but​‌ it is one possible​​ path towards the development​​ of new drug discovery​​​‌ and has been already‌ proposed and stated to‌​‌ be explored, in particular​​ by members of AIstroSight​​​‌ (B. Vidal, L. Zimmer)‌ 72, 59.‌​‌ Our objective here is​​ to build realistic mechanistic​​​‌ models of GPCR-based cell‌ signaling in the neuronal‌​‌ intracellular space. We plan​​ then to use this/these​​​‌ models to propose molecular‌ mechanisms to explain the‌​‌ experimentally observed biases. A​​ first idea to explore​​​‌ is the hypothesis of‌ a local subcellular compartmentalization‌​‌ of the signaling molecules​​ over and close to​​​‌ the cell membrane (so-called‌ nanodomains). Experimental validation of‌​‌ the main model predictions​​ is then to be​​​‌ performed using brain imaging‌ modalities available at CERMEP‌​‌ (TEP, MRI, fUS).

-​​ Synaptic plasticity: Synaptic​​​‌ plasticity, the long-term adaptation‌ of the efficacy of‌​‌ a synapse according to​​ the activity of the​​​‌ neurons and astrocytes composing‌ this synapse, is thought‌​‌ to underly learning and​​ memory at the cellular​​​‌ scale 68. We‌ have been enjoying a‌​‌ very fruitful collaboration with​​ Laurent Venance's lab (INSERM​​​‌ U1050, CIRB, Collège de‌ France, Paris) on the‌​‌ subcellular mechanisms at play​​ in learning and memory​​​‌ formation by synaptic plasticity‌ 36, 38,‌​‌ 39, 37,​​ 45, 74,​​​‌ 49. Current work‌ focuses on the control‌​‌ of synaptic plasticity mechanisms​​ by endocannabinoids and its​​​‌ implication in fast learning‌ and on metabolic regulation‌​‌ of synaptic plasticity by​​ astrocytes. This collaboration is​​​‌ funded by ongoing ANR‌ project EngFlea (see below).‌​‌

- Calcium signaling in​​ astrocytes: Calcium signaling​​​‌ in the terminal branchlets‌ of astrocytes is thought‌​‌ to be crucial for​​ astrocytic functions and neuron-astrocyte​​​‌ interactions 28. We‌ are studying the local‌​‌ dynamics of calcium signaling​​ in terminal branchlets of​​​‌ astrocytes and their interaction‌ with synaptic activity in‌​‌ collaboration with U. Valentin​​ Nägerl's lab (CNRS UMR​​​‌ 5297, Bordeaux) for experimental‌ (subcellular) data with supra-resolution‌​‌ microscopy 40. Recently,​​ a collaboration with Erik​​​‌ de Schutter's lab (Okinawa‌ Institute of Science and‌​‌ Technology, Japan) has also​​ been set up to​​​‌ develop new efficient modelling‌ tools (stochastic reaction-diffusion systems)‌​‌ in realistic 3D geometric​​ meshes based on the​​​‌ simulation framework they develop,‌ STEPS 27, 41‌​‌.

- Multiscale modelling​​ of the effects of​​​‌ a candidate drug on‌ neurons and astrocytes:‌​‌ To model the cellular​​ effect of a candidate​​​‌ drug, the main molecular‌ systems impacted by the‌​‌ drug are isolated from​​ the cellular signature data​​​‌ (from e.g., transcriptomics) and‌ literature exploration. Imaging data,‌​‌ by specifying the brain​​ areas and structures mainly​​​‌ targeted by the candidate‌ drug, helps refine these‌​‌ models using specific parameters.​​ Whereas in the first​​​‌ models, the observation (and‌ modelling scale) corresponds to‌​‌ a subcellular domain (one​​ synapse, +/- a dendrite​​​‌ or the main astrocytic‌ process in the neighborhood),‌​‌ we search to progressively​​ scale up those mechanistic​​​‌ models from the intracellular‌ scale of a single‌​‌ cell to the scale​​ of a population of​​​‌ interacting brain cells, neurons‌ and astrocytes. To do‌​‌ so, we explore model​​​‌ simplification /reduction methods, including​ those combining machine learning​‌ and dynamical systems modelling​​ (see below). In the​​​‌ long run, this large-scale​ mathematical model will produce​‌ a digital twin of​​ the pathology that will​​​‌ allow us to explain​ why the candidate drug​‌ has a positive effect​​ on the disease. Calibration​​​‌ is based on fUS​ and fMRI imaging data​‌ in rodents obtained in​​ the framework of the​​​‌ NI2D LabCom. This data​ provides us with quantitative​‌ measurements of the effects​​ of microscopic perturbations by​​​‌ pharmacological agents or by​ external stimuli (e.g., visual)​‌ on the variation, correlation​​ and spreading of cortical​​​‌ activity over the whole​ brain.

- Astrocyte roles​‌ on brain imaging signals​​: Although it is​​​‌ now widely accepted that​ astrocytes play a role​‌ in brain processes and​​ pathologies, the exact perimeters​​​‌ of their roles remain​ to be delimited. For​‌ instance, variations of the​​ signals measured by brain​​​‌ imaging methods (fMRI, PET,​ fUS) are still largely​‌ interpreted as variations of​​ neuronal activity. Available experimental​​​‌ data however indicate that​ astrocytes also impact those​‌ signals but it not​​ clear yet how they​​​‌ do it. A precise​ and quantitative answer to​‌ this question would allow​​ us to use brain​​​‌ imaging to monitor not​ only the local activity​‌ of the neurons but​​ also of the astrocytes.​​​‌ Such a feature would​ be precious in our​‌ framework of astrocyte pharmacology​​ but it demands the​​​‌ development of new mathematical​ models. Existing models of​‌ fMRI signals, for instance,​​ are either too crude​​​‌ to incorporate a separate​ astrocyte action (balloon models​‌ 47) or are​​ limited to the role​​​‌ of astrocytes as energy​ suppliers of the neurons​‌ (astrocyte-neuron lactate shuttle 53​​). Our objective here​​​‌ is to start from​ a microscopic and mechanistic​‌ model of neuron-astrocyte-blood vessel​​ interactions and use multi-scale​​​‌ modelling methodologies to obtain​ a large-scale model of​‌ astrocyte-neuron impact on a​​ subset of brain imaging​​​‌ technics (fMRI, fUS), with​ explicit parametrization of local​‌ neuronal and astrocyte activities.​​ Here again these models​​​‌ are calibrated using fUS​ and fMRI imaging data​‌ in rodents, in particular​​ using pharmacological agents that​​​‌ are known to specifically​ silence the astrocytic population​‌ or a neuronal population​​ in a given brain​​​‌ area. A crucial step​ is the development of​‌ a detailed, microscopic model​​ of the astrocyte endfeet,​​​‌ the specialized astrocyte processes​ that respond to and​‌ control local vascular diameters.​​ This model will provides​​​‌ us with causal mechanisms​ able to interlink neuronal​‌ electrical activity, astrocytic calcium​​ activity and local blood​​​‌ flow. It is to​ be seen as a​‌ first stage towards understanding​​ the implication of astrocytes​​​‌ in variations of neuroimaging​ signals.

Methodological challenges: The​‌ biological systems to be​​ considered to explain drug​​​‌ effects on pathologies are​ not only very complex​‌ but also only partly​​ understood by neurobiologists themselves.​​​‌ Therefore, the available biological​ knowledge on these systems​‌ is constantly evolving. Since​​ we cannot know in​​​‌ advance what systems are​ affected by the candidate​‌ drug, a major difficulty​​ for modelers is preparedness,​​ i.e. maintain a level​​​‌ of expertise on the‌ biology and modelling state-of-the-art‌​‌ of a wide range​​ of those systems. This​​​‌ is the reason why‌ the first three projects‌​‌ above are crucial to​​ the success of our​​​‌ proposal.

The most important‌ challenge we face is‌​‌ that of multiscale: causal​​ data are mainly molecular​​​‌ but many observations are‌ macroscopic (e.g. brain imaging).‌​‌ Traditionally, linking these two​​ scales requires the development​​​‌ of new theories (e.g.,‌ homogenization, population boundaries, etc.),‌​‌ a slow and rather​​ hazardous process. The availability​​​‌ of more and more‌ important computing resources allows‌​‌ to consider "brute force"​​ approaches in which all​​​‌ scales of time and‌ space are numerically simulated‌​‌ (cf Blue Brain Project​​). But the results​​​‌ are often as difficult‌ to interpret as the‌​‌ animal experiments that these​​ simulations emulate. Instead we​​​‌ consider recent advances in‌ hybrid digital-AI systems (physics-informed‌​‌ neural networks 64),​​ in particular equation discovery​​​‌ methodologies 63. These‌ methods usually use sparse‌​‌ regression techniques to select​​ in a library of​​​‌ nonlinear terms and operators,‌ those that, when composed,‌​‌ provide the best description​​ of the data 32​​​‌. Our idea is‌ to generate a large‌​‌ number of numerical simulations​​ at the microscopic scale​​​‌ of the kinetics of‌ the biochemical reactions concerned,‌​‌ for example by the​​ spatial Gillespie algorithm, and​​​‌ then to aggregate them‌ at a higher spatio-temporal‌​‌ scale using, for example,​​ averages at a space​​​‌ and time grain much‌ higher than the spatio-temporal‌​‌ resolution of the initial​​ microscopic simulations. The idea​​​‌ is then to use‌ equation-discovery algorithms to infer‌​‌ a set of differential​​ equations (and associated parameters)​​​‌ capable of describing these‌ higher scale space-time kinetics.‌​‌ The resulting reduced model​​ is then replicated in​​​‌ each cell of the‌ cell population model. If‌​‌ successful, this model reduction​​ process can even be​​​‌ reiterated at the upper‌ scale to simulate the‌​‌ effect of the molecule​​ on large brain areas.​​​‌ Of course, the risky‌ and difficult nature of‌​‌ this objective makes it​​ a long-term goal. If​​​‌ need be, alternative meta-modelling‌ technics will also be‌​‌ considered when applicable (RKHS,​​ model-order reduction).

3.2 Multi-patient​​​‌ query for care pathway‌ characterization and clinical trials‌​‌

Real-world data, especially the​​ data that is routinely​​​‌ collected by hospitals (medical‌ reports, hospital records…), provides‌​‌ rich information about possible​​ links between patient information​​​‌ (demographic, pathological, life style),‌ drug exposures and health‌​‌ events. In the context​​ of drug development, this​​​‌ data can be useful‌ at three stages. During‌​‌ the search for a​​ new drug, they can​​​‌ be used to enrich‌ cell culture data or‌​‌ imaging data. In this​​ case, one can query​​​‌ patients that have been‌ treated for the pathology‌​‌ in question and integrate​​ their clinical data in​​​‌ the in-silico screening. This‌ approach is presented below‌​‌ in the framework of​​ data integration.

Electronic Health​​​‌ Records query algorithms:

Efficient‌ patient query can also‌​‌ be used at the​​ very initial stage of​​​‌ drug discovery: the assessment‌ of the feasibility of‌​‌ drug development projects. Indeed,​​​‌ part of the pathologies​ we target are rare​‌ diseases. In this context,​​ one has to make​​​‌ sure at the very​ early stages that the​‌ pathology in question is​​ not so rare that​​​‌ the number of patients​ is too low to​‌ allow clinical trials, or​​ that its description in​​​‌ terms of physiopathology is​ mature enough for the​‌ clinician to be able​​ to diagnose it with​​​‌ good probability. We thus​ develop patient query algorithms​‌ on clinical data from​​ hospitals (electronic health records,​​​‌ EHR), in particular of​ the HCL, that allow​‌ us to characterize the​​ care pathways of the​​​‌ patients before and after​ diagnosis. They provides us​‌ with answers to many​​ questions related to the​​​‌ clinical picture of the​ pathology, its genetic underpinnings,​‌ its prevalence rate, the​​ typical care pathway of​​​‌ a patient with this​ pathology, the delay of​‌ diagnostic, the frequency of​​ diagnostic errors etc. Answers​​​‌ to these questions are​ crucial to determine early​‌ on whether the drug​​ discovery project is feasible.​​​‌ We aim at developping​ query algorithms and software​‌ pipelines for EHR that​​ can provide us with​​​‌ tools able to answer​ these questions efficiently.

Efficient​‌ EHR query algorithms are​​ also very useful at​​​‌ the final stage of​ the clinical trial itself​‌ (Fig. 2), where​​ they can be used​​​‌ to finely select what​ patients should be integrated​‌ in the trial. Indeed,​​ a major change of​​​‌ paradigm in medicine in​ recent years is the​‌ acceptation that the response​​ of a group of​​​‌ patients to a drug​ treatment exhibits strong variability.​‌ The source of this​​ variability is diverse 73​​​‌. The definition of​ the pathology itself as​‌ a unique coherent class​​ can be misleading and​​​‌ actually incorporate a range​ of different sub-classes of​‌ pathologies/disorders. The response to​​ a drug also depends​​​‌ on how the patient's​ body affects the drug​‌ before it reaches its​​ target organ/cells (pharmacokinetics). At​​​‌ the cellular level, the​ response can also vary​‌ because of inter-individual differences​​ in gene sequence and​​​‌ receptor/protein structure (pharmacodynamics). Therefore,​ individual drug responses depend​‌ on the patient genes​​ (pharmacogenetics) but also on​​​‌ more social factors (age,​ sex, anterior medical record,​‌ lifestyle, habits, exposure to​​ pollution…). In any case,​​​‌ the strength of this​ variability is believed to​‌ be a major cause​​ of failure for clinical​​​‌ trials, in particular in​ neurology and psychiatry 60​‌, 54. The​​ goal of “stratified medicine”​​​‌ in this perspective is​ to subdivide the available​‌ group of patients into​​ a number of subgroups​​​‌ so that the response​ of each subgroup is​‌ less variable than the​​ whole 42. Our​​​‌ objective is to develop​ computational tools and software​‌ packages able to stratify​​ hospital data to assist​​​‌ in the selection of​ patients to be included​‌ in an evaluation protocol​​ for a clinical trial​​​‌ or the building of​ a research cohort.

Computational​‌ phenotyping:

The task of​​ querying patients according to​​​‌ a predefined criterion from​ a large population of​‌ EHRs is sometimes referred​​ to as “computational phenotyping”​​ 65. It remains​​​‌ a time-consuming and challenging‌ task with complex criteria‌​‌ because the query is​​ to be addressed within​​​‌ multiple document types and‌ across multiple data points,‌​‌ in EHRs that usually​​ comprise both structured and​​​‌ unstructured data. The computational‌ challenges raised by patient‌​‌ query with complex criteria​​ are therefore considerable (integration,​​​‌ query, analysis, privacy). Software‌ tools (i2b2, ACE 33‌​‌) have been proposed​​ to query patients for​​​‌ cohorts or clinical trials‌ based on EHRs but‌​‌ they can hardly be​​ used by most of​​​‌ the physicians because they‌ require advanced knowledge of‌​‌ the data in computer​​ science terms (format, encoding).​​​‌ Moreover, our objective is‌ to provide clinicians with‌​‌ tools able to manipulate​​ these complex data together​​​‌ with medical concepts (e.g.,‌ exposure to a drug,‌​‌ treatment, or occurrence of​​ a pathology). Data abstraction​​​‌ capabilities must therefore be‌ integrated to automatically enrich‌​‌ the data using phenotype​​ libraries that can be​​​‌ intuitively mobilized by the‌ clinician. In analogy with‌​‌ bioinformatics workflows, we create​​ workflows for computational phenotyping.​​​‌

In cases where we‌ already know how to‌​‌ stratify, the issue is​​ not a learning problem​​​‌ but rather a query‌ problem. On the other‌​‌ hand, when this is​​ not the case, we​​​‌ have to develop methods‌ to build these homogeneous‌​‌ subgroups, and in this​​ case it is a​​​‌ question of (unsupervised) learning:‌ the training criterion becomes‌​‌ a measure of cluster​​ homogeneity. Two competing approaches​​​‌ can be thought of‌ in order to create‌​‌ the building blocks of​​ the workflow: 1) machine​​​‌ learning approaches that allow‌ the construction of abstract‌​‌ patient phenotypes from massive​​ data; 2) approaches inspired​​​‌ by both timed systems‌ modeling and knowledge reasoning‌​‌ that rely on formal​​ descriptions of computational phenotypes​​​‌ to enrich the data.‌ The interest of formal‌​‌ descriptions is to be​​ able to represent the​​​‌ whole data transformation in‌ a formal way. This‌​‌ abstract representation of the​​ construction of a cohort​​​‌ facilitates its understanding by‌ users and its reproducibility‌​‌ (FAIR principle). On the​​ other hand, they allow​​​‌ again to exploit intimately‌ the formalized knowledge of‌​‌ the domain, but they​​ also become objects that​​​‌ can be manipulated by‌ reasoning tools. The use‌​‌ of semantic web technologies​​ can therefore be an​​​‌ interesting tool for representing‌ data, knowledge and their‌​‌ processing in order to​​ propose query tools that​​​‌ guide the clinician through‌ the knowledge.

Methodological challenge:‌​‌

The challenge is to​​ make these formal descriptions​​​‌ highly expressive and to‌ ensure efficient processing of‌​‌ massive data. On the​​ long run, we plan​​​‌ to take inspiration from‌ the approach called “Ontology-Mediated‌​‌ Query Answering” which consists​​ in using ontologies to​​​‌ mediate the query of‌ a database by ontologies‌​‌ 31. In this​​ context, a computational phenotype​​​‌ is seen as a‌ query. The difficulties encountered‌​‌ with observational data is​​ the semantic gap between​​​‌ the available data and‌ the medical concepts that‌​‌ are interesting to manipulate.​​ This gap may be​​​‌ bridged by automatic reasoning‌ that exploits expert knowledge‌​‌ to relate different abstraction​​​‌ levels.

Since computational phenotypes​ are difficult to formalize,​‌ the challenge is to​​ support clinicians in defining​​​‌ them. In other words,​ the challenge becomes to​‌ abstract phenotypes from clinical​​ data. We plan to​​​‌ combine automatic reasoning methods​ and data analysis. The​‌ first research direction we​​ propose is the exploration​​​‌ of a symbolic approach​ parallel to the work​‌ by Tijl de Bie​​ 30 or by Silberschatz​​​‌ 67 on the notion​ of "subjective measure of​‌ interestingness". This approach was​​ developed to identify user-relevant​​​‌ statistical analysis results by​ means of a statistical​‌ model to evaluate the​​ novelty of the extracted​​​‌ patterns (a priori knowledge​ model). Symbolic approaches can​‌ be combined in a​​ similar way by using​​​‌ symbolic data analysis methods​ such as pattern mining,​‌ and by relying on​​ formal models of the​​​‌ system as a priori​ knowledge. Patterns that are​‌ not "explainable" by the​​ formal model are potentially​​​‌ new or of particular​ interest to the user​‌ and will thus be​​ extracted. This approach offers​​​‌ an original entry point​ to deeply integrate knowledge-based​‌ reasoning into pattern extraction​​ methods. The research challenge​​​‌ here lies in combining​ formalized knowledge with experimental​‌ data. It may be​​ implemented using the declarative​​​‌ pattern mining paradigm, that​ uses solvers to address​‌ the pattern mining task.​​ The proofs of concept​​​‌ on the notion of​ novelty will open the​‌ way to more complex​​ reasoning such as planning​​​‌ that can be used​ to integrate complex behaviors​‌ in biological systems, such​​ as interaction networks. The​​​‌ second research direction we​ propose is based on​‌ recent machine learning techniques.​​ Unsupervised ML has been​​​‌ applied to patient phenotyping,​ i.e. the discovery of​‌ phenotypes from EHR data,​​ including temporal phenotyping 75​​​‌. Our objective is​ to combine such kinds​‌ of algorithm with semantic​​ knowledge to guide the​​​‌ discovery toward meaningful computational​ phenotypes. Indeed data embedding​‌ techniques can integrate ontologies​​ to enhance data semantics​​​‌ 46.

On the​ long run, the methods​‌ developed above may be​​ reunified to address the​​​‌ problem of drug discovery​ at both the biological​‌ and the body scale.​​ This justifies the coherence​​​‌ of the methodological approaches​ (Semantic Web and machine​‌ learning) that are developed​​ in the two objectives.​​​‌

4.1​ Targeted Pathologies

The list​‌ of pathologies that are​​ of interest for Theranexus​​​‌ in the framework of​ AIstroSight is given in​‌ the stand-alone “convention d'équipe-projet​​ commune” of AIstroSight. It​​​‌ comprises roughly 30 rare​ diseases of the central​‌ nervous system, including lysosomal​​ pathologies, neurological genetic diseases,​​​‌ rare diseases due to​ $\alpha$-synuclein accumulation or​‌ rare demyelinating pathologies. This​​ list may be updated,​​​‌ subject to the prior​ joint written consent of​‌ AIstroSight partners. For the​​ pathologies in this list,​​​‌ a specific regimen is​ defined in terms of​‌ IP and legal affairs.​​ AIstroSight members are allowed​​​‌ to work on pathologies​ outside this list without​‌ any restriction but with​​ a different legal regimen​​​‌ vis-à-vis Theranexus. In agreement​ with the company, we​‌ have selected two pathologies​​ from this list as​​ priority objectives, on which​​​‌ we will start our‌ work: Rett syndrome and‌​‌ Niemann-Pick Type C disease.​​ Both pathologies are neurodevelopmental​​​‌ diseases caused by mutations‌ in a single gene:‌​‌ mutations in methyl-CpG-binding protein​​ 2 (MeCP2) for Rett​​​‌ 34 and in NPC‌ for Niemann-Pick C 62‌​‌. MeCP2 mutation in​​ Rett syndrome causes slowed​​​‌ brain growth, a progressive‌ loss of movement, motor‌​‌ control abilities and language​​ in the children and​​​‌ can also cause heavy‌ breathing problems, epileptic-like seizures‌​‌ or intellectual disabilities, among​​ others 52. NPC​​​‌ mutations in Niemann-Pick type‌ C causes accumulation of‌​‌ cholesterol and other fatty​​ acids inside lysosomes, including​​​‌ in brain cells. The‌ symptoms are highly variable,‌​‌ ranging from defects of​​ developmental and motor progression,​​​‌ difficulties in learning, speech‌ or swallowing, to cognitive‌​‌ impairment or psychiatric symptoms​​ 70. Restoring NPC​​​‌ expression in astrocytes significantly‌ increases survival of mice‌​‌ models of the disease,​​ suggesting that astrocyte dysfunction​​​‌ is involved in disease‌ progression 76. The‌​‌ two diseases also have​​ in common that they​​​‌ are rare neurological diseases‌ of children (frequency $\mathrm{1‌‌}/{10}^4$ and​​ $1/{10}^{\mathrm{5‌}}$, respectively) and that‌ there is no known‌​‌ effective treatment. This rarity​​ has strong consequences on​​​‌ the numerical tools that‌ can be used to‌​‌ find potential pharmacological targets.​​

4.2 In vitro and​​​‌ in vivo experimental models‌ of neurology diseases

Many‌​‌ of the diseases that​​ are of interest for​​​‌ AIstroSight are rare diseases.‌ This means that the‌​‌ volume of experimental data​​ and the basic understanding​​​‌ of the pathology at‌ the (sub-)cellular level may‌​‌ be too limited for​​ the machine learning or​​​‌ mechanistic modeling tools that‌ we plan to use.‌​‌ For example, it is​​ known that the NPC​​​‌ mutation in Niemann-Pick type‌ C induces morbid cholesterol‌​‌ accumulation in cells but​​ the molecular function of​​​‌ NPC in cholesterol metabolism‌ is not clearly understood‌​‌ 62. Similarly, MeCP2,​​ the gene mutated in​​​‌ Rett syndrome, is an‌ epigenetic regulatory factor (DNA‌​‌ methylation) whose mutation theoretically​​ impacts the expression of​​​‌ a large number of‌ genes but it is‌​‌ not clear which ones​​ are most involved in​​​‌ the symptoms of the‌ disease 52. Although‌​‌ molecular (omic) studies have​​ been published for both​​​‌ diseases 35, 66‌, their molecular contexts‌​‌ are still unclear.

Our​​ goal here is to​​​‌ generate additional preclinical molecular‌ and imaging data to‌​‌ better delineate the perturbations​​ that these diseases cause​​​‌ at a cellular and‌ tissue level. We introduce‌​‌ into cultured cells the​​ same deficits as those​​​‌ observed in patients. Transcriptomic‌ analysis of the effect‌​‌ of this manipulation gives​​ us information on the​​​‌ implicated molecular networks and‌ its major molecular consequences.‌​‌ In parallel, we induce​​ these same perturbations in​​​‌ vivo in rodents. Observing‌ these animals using brain‌​‌ imaging techniques (fMRI and​​ fUS, possibly PET) gives​​​‌ us a more macroscopic‌ view of the effect‌​‌ of the mutation (affected​​ brain areas, nature and​​​‌ amplitude of the modifications,‌ change in response to‌​‌ treatments or stimuli etc,​​​‌ see below).

Methodological challenges​: Developing experimental models​‌ of pathologies can be​​ a very difficult task​​​‌ for pathologies that are​ due to the conjunction​‌ of multiple factors, when​​ the molecular alterations at​​​‌ the origin of the​ pathologies have effects over​‌ a very large range​​ of cellular processes or​​​‌ when comparison of the​ phenotype of the experimental​‌ model with its human​​ counterpart is ill-defined (psychiatric​​​‌ diseases, for instance). To​ mitigate this risk, we​‌ develop experimental models only​​ for pathologies that are​​​‌ well-defined in molecular terms,​ like for Rett or​‌ Niemann-Pick type C for​​ a start. We use​​​‌ viral vector strategies (mostly​ shRNA-mediated gene silencing or​‌ possibly CRISPR-based gene editing​​ via adeno-associated viruses, AAV)​​​‌ to manipulate the sequence​ or expression of the​‌ target gene. We start​​ with cell lines that​​​‌ are easy to grow​ and analyze using omics​‌ approaches, and then use​​ neurons and astrocytes differentiated​​​‌ from human pluripotent stem​ cells. This approach is​‌ also used in vivo​​ by locally injecting the​​​‌ viral vector into a​ given brain region of​‌ an animal model, to​​ genetically modify a particular​​​‌ cell type by using​ a specific promoter. We​‌ should therefore be able​​ to control the area​​​‌ of the brain in​ which the genetic manipulation​‌ will be induced (e.g.​​ visual cortex or cerebellum)​​​‌ as well as the​ type of cells targeted​‌ (neurons vs. astrocytes, for​​ example). Of course, like​​​‌ all experimental models, each​ model taken separately has​‌ its limitations: the genes​​ expressed by cells in​​​‌ culture are not necessarily​ those expressed by these​‌ same cells in vivo,​​ the effects of gene​​​‌ silencing in a rodent​ are not necessarily transposable​‌ to humans, etc. However​​ our hypothesis is that​​​‌ by combining these different​ modalities and scales of​‌ data (see above), it​​ should be possible to​​​‌ better predict the effect​ of a potential treatment.​‌ The molecular and cellular​​ biology technologies to be​​​‌ mobilized here (in vitro​ and in vivo mutagenesis,​‌ cell culture, proteomics) are​​ tools routinely used by​​​‌ Theranexus. The expertise on​ the use of medical​‌ imaging to observe the​​ effects at the brain​​​‌ level is provided by​ CERMEP and benefits from​‌ the advances of the​​ NI2D LabCom.

4.3 Identification​​​‌ of multi-source multi-scale biomarkers​

Recently, Theranexus changed its​‌ pharmacological strategy, from a​​ strategy mainly based on​​​‌ the repositioning of pre-existing​ drugs to a technology​‌ based on antisense oligonucleotide​​ drugs. These technologies rely​​​‌ on the ability to​ design on demand short​‌ RNA sequences that specifically​​ bind the mRNA of​​​‌ a gene target, and​ knock it down after​‌ recognition by the RNase​​ H1 enzymes present in​​​‌ all cells, or modulate​ its translation or splicing​‌ via steric hindrance .​​ Pharmacological intervention thus consists​​​‌ in searching for a​ gene target able to​‌ correct the molecular perturbation​​ caused by the disease​​​‌ and to synthesize an​ antisense oligonucleotide able to​‌ specifically bind this target​​ gene. Note that the​​​‌ technology currently in clinical​ use does not (yet)​‌ provide ways to specifically​​ target a cell type​​ or a brain region.​​​‌

Our first objective is‌ to develop digital tools‌​‌ to model the molecular​​ networks perturbated by the​​​‌ pathology of interest, and‌ use this model to‌​‌ identify a gene or​​ protein in the network​​​‌ the modulation of which‌ would correct the perturbation‌​‌ caused by the pathology.​​ These models are based​​​‌ on molecular data, in‌ particular transcriptomics and metabolomics‌​‌ data. The set of​​ data includes data derived​​​‌ from cell cultures as‌ described above, that we‌​‌ augment with molecular data​​ from the literature related​​​‌ to the pathology or‌ more generic public, open‌​‌ access databases of transcriptomic​​ responses to perturbating molecules,​​​‌ like CMap or the‌ LINCS L1000 data repository‌​‌. The latter, for​​ instance currently includes the​​​‌ effect of close to‌ 40,000 small perturbating molecules‌​‌ on 12,000+ genes of​​ more than 200 cell​​​‌ types. We aggregate‌ these data and use‌​‌ them to infer the​​ gene interaction network, the​​​‌ metabolic network and/or the‌ signaling network impacted by‌​‌ the pathology. Metabolic networks​​ are important for instance​​​‌ for Niemann-Pick type C,‌ to conciliate perturbations of‌​‌ the lipid metabolism with​​ those of the gene​​​‌ expression network. This provides‌ us with a view‌​‌ of the pathology at​​ the molecular scale.

Integration​​​‌ of neuroimaging data:

A‌ major objective of AIstroSight‌​‌ is to augment these​​ molecular data with medical​​​‌ data, in particular brain‌ imaging data and hospital‌​‌ data. We complement molecular​​ data with data coming​​​‌ from the analysis of‌ brain imaging (fMRI, PET,‌​‌ functional ultrasound brain imaging)​​ i.e. with functional networks​​​‌ between brain areas targeted‌ by the molecule or‌​‌ quantitative measures of radioligand​​ binding. Most of this​​​‌ imaging is done in‌ rodents (preclinic, see above)‌​‌ but a subset of​​ human imaging data is​​​‌ also used. These imaging‌ data are obtained by‌​‌ our collaborators from the​​ CERMEP platform.

These different​​​‌ neuroimaging methods provide meaningful‌ and complementary information for‌​‌ understanding the functional or​​ molecular effects of drugs​​​‌ in the brain:

• Positron‌ emission tomography (PET) enables‌​‌ to visualize and measure​​ the concentration of a​​​‌ specific radiotracer, with nanomolar‌ to picomolar sensitivity. Numerous‌​‌ brain PET radiotracers have​​ been developed over the​​​‌ years, enabling to study‌ various molecular processes (such‌​‌ as the synthesis and​​ release of endogenous neurotransmitters,​​​‌ the density of receptors,‌ transporters or proteins aggregates,‌​‌ neuroinflammation and cerebral metabolism)​​ both in animals and​​​‌ humans in a non-invasive‌ way. We especially focus‌​‌ on the measurement of​​ cerebral glucose consumption using​​​‌ [18F]FDG, a radiolabeled glucose‌ analog, while also taking‌​‌ advantage of the rich​​ information that can be​​​‌ obtained using other brain‌ radiotracers when needed. Indeed,‌​‌ the asset of [18F]FDG-PET​​ imaging is to be​​​‌ relevant for virtually all‌ drugs that are expected‌​‌ to be active in​​ the brain, since the​​​‌ cerebral glucose uptake is‌ related to neuron and‌​‌ astrocyte activities. In addition,​​ this neuroimaging technique can​​​‌ be used in awake‌ freely-moving animals, getting rid‌​‌ of the anesthesia or​​ stress confound in preclinical​​​‌ studies.
• Functional magnetic resonance‌ imaging (fMRI) enables to‌​‌ follow the dynamic changes​​​‌ in the BOLD signal​ (for “blood-oxygen level dependent”),​‌ which is also related​​ to the brain activity,​​​‌ in a non-invasive way.​ It has a high​‌ temporal resolution (2 to​​ 3 seconds) as compared​​​‌ to PET imaging (several​ minutes) and therefore can​‌ be used either to​​ measure the time series​​​‌ of BOLD signal changes​ after injection of a​‌ drug (called “pharmaco-MRI”) or​​ the changes in functional​​​‌ connectivity occurring after neuropharmacological​ stimulation. Functional connectivity is​‌ defined as the correlation​​ in activity over time​​​‌ between different brain areas;​ this concept has largely​‌ become prominent in neurosciences​​ over the years, and​​​‌ it is known to​ be modified in many​‌ physiological or pathological states.​​ fMRI can be used​​​‌ in animals and humans,​ similarly to PET, but​‌ usually requires anesthesia for​​ preclinical applications.
• Functional ultrasound​​​‌ imaging (fUS) is a​ much more recent imaging​‌ technology. It provides access​​ to the dynamic measurement​​​‌ of cerebral blood volumes​ changes, which are more​‌ straightforwardly related to the​​ brain activity as compared​​​‌ to the BOLD signal.​ Moreover, its spatial, temporal​‌ resolution and sensitivity are​​ unmatched by the previous​​​‌ techniques, and it can​ be applied to freely-moving​‌ awake animals in real-time,​​ in correlation with a​​​‌ particular behavior. However, it​ is currently limited to​‌ imaging in 2-dimensions (one​​ brain plane at a​​​‌ time) and is mostly​ suitable for animals. Therefore,​‌ fUS imaging is a​​ complementary way to study​​​‌ brain activity in small​ animals in the context​‌ of preclinical neuropharmacology.

Integration​​ of clinical data:

We​​​‌ also plant ot integrate​ hospital data from the​‌ Hospices Civils de Lyon​​ according to availability and​​​‌ pathologies. Hospital data provide​ access to rich information​‌ on possible links between​​ patient information (demographic, pathological),​​​‌ drug exposures, health events​ or biological sample analysis​‌ (e.g., blood markers). Our​​ goal is to integrate​​​‌ brain imaging and hospital​ data with cellular signatures​‌ to enrich them with​​ information at the individual​​​‌ scale in a form​ that can be analyzed​‌ with machine learning (clustering,​​ classification) or data mining​​​‌ (pattern matching) methods.

Methodological​ challenges:

A first challenge​‌ resides in the nature​​ of action of antisense​​​‌ oligonucleotides, that often work​ by knock down/loss of​‌ function. It is not​​ straightforward to design such​​​‌ a strategy in the​ case of a pathology​‌ that is due to​​ a mutation that already​​​‌ suppressed the effect of​ a gene. That is​‌ precisely where numerical models​​ of the involved gene​​​‌ expression and metabolic networks​ are important because they​‌ can be systematically assessed​​ for the effect of​​​‌ gene suppression, thus providing​ a quick in silico​‌ screening of the potential​​ targets. However, part of​​​‌ this program implies typical​ bioinformatics processing steps: analysis​‌ of transcriptomic networks, network​​ reconstruction, conciliation between transcriptomic​​​‌ and metabolic networks… We​ currently do not have​‌ this expertise in the​​ team. Therefore we leverage​​​‌ collaborations with local experts​ of the field to​‌ get the necessary operational​​ knowledge, including experts of​​​‌ brain transcriptomics analysis (MeLiS​ lab in Lyon).

Another​‌ difficulty lies in the​​ heterogeneity of these multiscale​​ data, their highly categorical​​​‌ character, the large dimension‌ of the corresponding variable‌​‌ space and often, the​​ small number of observations.​​​‌ Moreover, cellular signature data‌ are intrinsically very noisy‌​‌ and can have low​​ reproducibility 56, a​​​‌ caveat that feature selection‌ may improve, at least‌​‌ in part 43.​​ Class imbalance can also​​​‌ be strong. Finally, each‌ type of available observation‌​‌ (molecular networks, imaging, hospital)​​ gives a partial, fragmented​​​‌ and incomplete view of‌ an abstract complex biological‌​‌ system. This is a​​ partial view because each​​​‌ type of observation provides‌ data at a given‌​‌ spatio-temporal scale, for a​​ certain locus. This is​​​‌ a fragmented view because‌ the data will be‌​‌ collected from different patients,​​ and even from very​​​‌ different living systems (cell‌ cultures, animals, patients). Each‌​‌ patient contributes to the​​ description of the abstract​​​‌ system on only few‌ types of observations. This‌​‌ is incomplete because there​​ will be many gaps​​​‌ to bridge the different‌ kinds of information related‌​‌ to functioning of the​​ studied biological system.

To​​​‌ reach our objective, we‌ explore the use of‌​‌ Semantic Web (SW) formalism​​ which attracts a lot​​​‌ of interest in bioinformatics,‌ to formalize knowledge and‌​‌ data. Data are observations​​ of biological systems acquired​​​‌ within controlled experiments or‌ in real life. Formalized‌​‌ knowledge is a representation​​ of facts and rules​​​‌ acquired in a scientific‌ domain, here medicine or‌​‌ life sciences. Applying machine​​ learning techniques on data​​​‌ supports knowledge discovery, but‌ it is only one‌​‌ particular source of knowledge.​​ The methodological challenge is​​​‌ first to formalize the‌ different types of available‌​‌ data within an abstract​​ model of the biological​​​‌ system, and to integrate‌ formalized knowledge in the‌​‌ model coming from medical​​ literature and our medical​​​‌ expertise, including imaging or‌ hospital data. By gathering‌​‌ a wide range of​​ formalized data and knowledge​​​‌ within the same tool,‌ we aim at creating‌​‌ a kind of abstract​​ numerical twin that may​​​‌ be queried to infer‌ new knowledge to assist‌​‌ drug design or drug​​ repositioning.

On the longer​​​‌ run, the second challenge‌ is to develop query‌​‌ answering at the abstract​​ level but based on​​​‌ fragmented data. The objective‌ is to answer queries‌​‌ about the numerical twin​​ by exploiting the data​​​‌ coming from multiple patients.‌ One of the difficulties‌​‌ is to detect groups​​ of patients whose numerical​​​‌ twins are “similar to‌ each other” (in a‌​‌ sense that remains to​​ be defined). Semantic Web​​​‌ offers a natural framework‌ for querying formalized data‌​‌ with multiple facets but​​ may be limited by​​​‌ the time-efficiency of the‌ query engines on a‌​‌ large number of patients.​​ In such a context,​​​‌ numerical approaches (embedding) is‌ more time-efficient but may‌​‌ lack accuracy. The challenge​​ is to construct numerical​​​‌ representations in order to‌ embed the data in‌​‌ a space in which​​ the distances are both​​​‌ efficient to compute and‌ semantically consistent with the‌​‌ applied notion of “similarity”.​​ Numerical machine learning techniques​​​‌ turn out to be‌ an interesting perspective to‌​‌ address this challenge 58​​​‌. Recent research on​ advanced machine learning, such​‌ as representation learning, offers​​ new perspectives to address​​​‌ our challenge. Our objective​ is to initiate collaborations​‌ with teams having strong​​ backgrounds in machine learning​​​‌ (e.g. Ockham Inria Team)​ to propose innovative solutions.​‌ Another important point is​​ the need for logic​​​‌ programming methodologies able to​ express complex queries, especially​‌ on heterogeneous or multimodal​​ data. For neuroimaging, the​​​‌ availability of Neurolang for​ logic programming with heterogeneous​‌ data or NeuroQuery for​​ query result consolidation based​​​‌ on automatic literature meta-analysis,​ for instance, should be​‌ very useful.

In most​​ of the cases, the​​​‌ methodologies that we use​ to reach the above​‌ objectives are related to​​ knowledge management/mining, formal reasoning,​​​‌ data mining or learning.​ Machine learning or deep​‌ learning approaches are probably​​ less useful here. The​​​‌ main reason is related​ to the volume of​‌ available data. For rare​​ disease like Niemann-Pick type​​​‌ C, for instance, the​ low prevalence means that​‌ 5 to 10 new​​ patients are diagnosed in​​​‌ France each year, a​ number too low for​‌ deep neural networks. However,​​ advances in transfer learning​​​‌ might be helpful here.​ For instance, a large​‌ number of brain pathologies​​ come with dysfunction of​​​‌ intracellular cholesterol metabolism and​ storage. This is for​‌ instance the case of​​ multiple sclerosis 57,​​​‌ for which large cohorts​ and databases are available​‌ worldwide. As a long​​ term project, an interesting​​​‌ idea will be to​ leverage the large volume​‌ of data on multiple​​ sclerosis to identify biomarkers​​​‌ of cholesterol dysfunction, e.g.,​ in neuroimaging, and use​‌ transfer learning to adapt​​ the network to Niemann-Pick​​​‌ type C patients.

5.1 Footprint of research​​ activities

The team aims​​​‌ to have as low​ a footprint as possible.​‌ For instance, we try​​ to maximize the lifespan​​​‌ of our computing equipments.​ Travel is done by​‌ train as much as​​ possible. We regret that​​​‌ the tools currently available​ are not designed to​‌ facilitate this type of​​ choices when traveling abroad,​​​‌ which results in a​ significant additional workload falling​‌ on researchers and research​​ team assistants. We believe​​​‌ there is significant room​ for improvement in strengthening​‌ the social and ecological​​ responsibility of Inria's research​​​‌ teams and its evaluation,​ and we hope that​‌ work will be done​​ to address this gap.​​​‌

5.2 Impact of research​ results

In a long​‌ term effort to move​​ the impact of our​​​‌ work closer to clincal​ applications, the team decided​‌ to relocate out of​​ the premises of Inria's​​​‌ Lyon research center. The​ whole team is located​‌ since March 2025 in​​ new offices in the​​​‌ basement of Lyon's hospital​ for neurology, in the​‌ Lyon Est - Bron​​ medical campus.

This move​​​‌ has already started to​ considerably amplify our interface​‌ with the local ecosystem​​ of academic and clinical​​​‌ research in neuroscience. Actually,​ the move allows us​‌ to interact more not​​ only with the clinical​​​‌ teams of the neurology​ hospital, it also considerably​‌ improved our visibility towards​​ all the clinical research​​ teams of the Hospices​​​‌ Civils de Lyon, Lyon's‌ University hospital. Further, it‌​‌ also very significantly improved​​ our engagement within the​​​‌ whole Lyon's research ecosystem,‌ since all the local‌​‌ neuroscience research units (CRNL,​​ SBRI, INMG,ISC-JM) are located​​​‌ within the Lyon Est‌ / Bron campus, a‌​‌ few steps away from​​ our new premise.

The​​​‌ impact of this move‌ will of course become‌​‌ fully visible in our​​ production after some delay,​​​‌ since research takes time.‌ But it has already‌​‌ been obvious in 2025​​ via e.g., the obtention​​​‌ of new grants with‌ local neuroscience research units‌​‌ (SBRI, ISC, MeLIS) and​​ clincal research teams (CICLY).​​​‌

7.1 Latest software developments‌

8.1 Investigating​​​‌ the ultrastructural properties of​ the endoplasmic reticulum in​‌ perisynaptic astrocytic processes and​​ its impact on signaling​​​‌

Participants: Audrey Denizot.​

Astrocytes recently emerged as​‌ key regulators of information​​ processing in the brain.​​​‌ Calcium signals in perisynaptic​ astrocytic processes (PAPs) notably​‌ allow astrocytes to fine-tune​​ neurotransmission at tripartite synapses.​​​‌ As most PAPs are​ below the diffraction limit,​‌ their content in calcium​​ stores and the contribution​​​‌ of the latter to​ astrocytic calcium activity is​‌ unclear.

In 2,​​ we reconstruct hippocampal tripartite​​​‌ synapses in 3D from​ a high resolution electron​‌ microscopy (EM) dataset and​​ find that 75% of​​​‌ PAPs contain some endoplasmic​ reticulum (ER), a major​‌ astrocytic calcium store. The​​ ER in PAPs displays​​​‌ strikingly diverse shapes and​ intracellular spatial distributions. To​‌ investigate the causal relationship​​ between each of these​​​‌ geometrical properties and the​ spatio-temporal characteristics of calcium​‌ signals, we implemented an​​ algorithm that generates 3D​​​‌ PAP meshes by altering​ the distribution of the​‌ ER independently from ER​​ and cell shape. Reaction-diffusion​​​‌ simulations in these meshes​ reveal that astrocyte activity​‌ is governed by a​​ complex interplay between the​​​‌ location of calcium channels,​ ER surface-volume ratio and​‌ spatial distribution. In particular,​​ our results suggest that​​​‌ ER-PM contact sites can​ act as local signal​‌ amplifiers if equipped with​​ IP3R clusters but attenuate​​​‌ PAP calcium activity in​ the absence of clustering.​‌ This study sheds new​​ light on the ultrastructural​​ basis of the diverse​​​‌ astrocytic calcium microdomain signals‌ and on the mechanisms‌​‌ that regulate neuron-astrocyte signal​​ transmission at tripartite synapses.​​​‌

8.2 Evaluating PDE discovery‌ methods for multiscale modeling‌​‌ of biological signals

Participants:​​ Andréa Ducos, Audrey​​​‌ Denizot, Thomas Guyet‌.

Biological systems are‌​‌ non-linear, include unobserved variables​​ and the physical principles​​​‌ that govern their dynamics‌ are partly unknown. This‌​‌ makes the characterization of​​ their behavior very challenging.​​​‌ Notably, their activity occurs‌ on multiple interdependent spatial‌​‌ and temporal scales that​​ require linking mechanisms across​​​‌ scales. To address the‌ challenge of bridging gaps‌​‌ between scales, we leverage​​ partial differential equations (PDE)​​​‌ discovery. PDE discovery suggests‌ meso-scale dynamics characteristics from‌​‌ micro-scale data.

In 12​​, we present our​​​‌ framework combining particle-based simulations‌ and PDE discovery and‌​‌ conduct preliminary experiments to​​ assess equation discovery in​​​‌ controlled settings. We evaluate‌ five state-of-the-art PDE discovery‌​‌ methods on particle-based simulations​​ of calcium diffusion in​​​‌ astrocytes. The performances of‌ the methods are evaluated‌​‌ on both the form​​ of the discovered equation​​​‌ and the forecasted temporal‌ variations of calcium concentration.‌​‌ Our results show that​​ several methods accurately recover​​​‌ the diffusion term, highlighting‌ the potential of PDE‌​‌ discovery for capturing macroscopic​​ dynamics in biological systems​​​‌ from microscopic data.

8.3‌ Enhancing Fluorescence Correlation Spectroscopy‌​‌ with machine learning to​​ infer anomalous molecular motion​​​‌

Participants: Nathan Quiblier,‌ Jan-Michael Rye, Hugues‌​‌ Berry.

The random​​ motion of molecules in​​​‌ living cells has consistently‌ been reported to deviate‌​‌ from standard Brownian motion,​​ a behavior coined as​​​‌ “anomalous diffusion”. To study‌ this phenomenon in living‌​‌ cells, Fluorescence Correlation Spectroscopy​​ (FCS) and Single-Particle Tracking​​​‌ (SPT) are the two‌ main methods of reference.‌​‌ In opposition to SPT,​​ FCS with its classical​​​‌ analysis methodology cannot consider‌ models of motion for‌​‌ which no analytical expression​​ of the auto-correlation function​​​‌ is known. This excludes‌ for instance anomalous Continuous-Time‌​‌ Random Walks (CTRW) and​​ Random Walk on fractal​​​‌ (RWf). Moreover, the whole‌ acquisition sequence of the‌​‌ classical FCS methodology takes​​ several tens of minutes.​​​‌

In 9, we‌ proposed a new analysis‌​‌ approach that frees FCS​​ of these limitations. Our​​​‌ approach associates each individual‌ FCS recording with a‌​‌ vector of features based​​ on an estimator of​​​‌ the auto-correlation function and‌ uses machine learning to‌​‌ infer the underlying model​​ of motion and to​​​‌ estimate the values of‌ the motion parameters. Using‌​‌ simulated recordings, we show​​ that this approach endows​​​‌ FCS with the capacity‌ to distinguish between a‌​‌ range of standard and​​ anomalous random motions, including​​​‌ CTRW and RWf. Our‌ approach exhibits performances comparable‌​‌ to the best-in-class state-of-the-art​​ algorithms for SPT and​​​‌ can be used with‌ a range of FCS‌​‌ setup parameters. Since it​​ can be applied on​​​‌ individual recordings of short‌ duration, we show that‌​‌ with our method, FCS​​ can be used to​​​‌ monitor rapid changes of‌ the motion parameters. Finally,‌​‌ we apply our method​​ on experimental FCS recordings​​​‌ of calibrated fluorescent beads‌ in increasing concentrations of‌​‌ glycerol in water. Our​​​‌ results accurately predict that​ the beads follow Brownian​‌ motion with a diffusion​​ coefficient and anomalous exponent​​​‌ which agree with classical​ predictions from Stokes-Einstein law​‌ even at large glycerol​​ concentrations. Taken together, our​​​‌ approach significantly augments the​ analysis power of FCS​‌ to capacities that are​​ similar to state-of-the-art SPT​​​‌ approaches.

This work was​ carried out in the​‌ framework of the ANR​​ project ABC4M (see 10.3​​​‌).

8.4 Single-cell multi​ omics data integration for​‌ gene regulatory network inference​​

Participants: Thibaut Peyric,​​​‌ Thomas Guyet.

In​ 14, we presented​‌ a novel three-state gene​​ expression model designed to​​​‌ elucidate the underlying mechanisms​ of mRNA transcription and​‌ its regulation. This is​​ a collaborative work between​​​‌ three Inria's team (Biotic,​ Music and AIstroSight) which​‌ has been awarded by​​ the best paper of​​​‌ the CMSB conference.

Our​ model incorporates gene regulatory​‌ processes by explicitly including​​ a transcription factor-bound state,​​​‌ thereby capturing the dynamic​ interplay between transcription activation​‌ and chromatin dynamics. We​​ fit the model to​​​‌ paired single-cell ATAC-seq and​ single-cell RNA-seq data, as​‌ these data give us​​ simultaneous information on a​​​‌ gene's transcriptional state and​ its accompanying chromatin state.​‌ Working at the pseudo-bulk​​ level, we extract biologically​​​‌ meaningful high-level descriptors from​ homogeneous cell (sub)populations, such​‌ as the mean and​​ variance of gene expression​​​‌ as well as the​ fraction of accessible chromatin.​‌ Crucial to the computational​​ feasibility of our approach,​​​‌ these descriptors can be​ analytically related to our​‌ model parameters.

The model​​ parameters reveal a small​​​‌ number of distinct expression​ strategies among gene clusters,​‌ providing data-driven novel insight​​ into context-dependent regulation of​​​‌ gene expression.

8.5 Vascular​ Segmentation of Functional Ultrasound​‌ Images using Deep Learning​​

Participants: Hana Sebia,​​​‌ Thomas Guyet, Hugues​ Berry.

fUS is​‌ a non invasive imaging​​ method that measures changes​​​‌ in cerebral blood volume​ (CBV) with high spatio-temporal​‌ resolution. It is a​​ recent technique used in​​​‌ premedical studies for instance​ to analyse the effects​‌ of drugs on brain​​ activity. In such purpose,​​​‌ it is then important​ to distinguish arterial flow​‌ from venuous flow. However,​​ distinguishing arterioles from venules​​​‌ in fUS is challenging​ due to opposing blood​‌ flow directions within the​​ same pixel.

In 11​​​‌, 24, we​ introduce the first deep​‌ learning-based application for fUS​​ image segmentation, capable of​​​‌ differentiating signals based on​ vertical flow direction (upward​‌ vs. downward), leveraging annotations​​ derived from Ultrasound Localization​​​‌ Microscopy (ULM) paired images,​ and enabling dynamic CBV​‌ quantification. In the cortical​​ vasculature, this distinction in​​​‌ flow direction provides a​ proxy for differentiating arteries​‌ from veins.

We evaluate​​ various UNet architectures on​​​‌ fUS images of rat​ brains, achieving competitive segmentation​‌ performance, with 90% accuracy,​​ a 71% F1 score,​​​‌ and an IoU of​ 0.59, using only 100​‌ temporal frames from a​​ fUS stack. These results​​​‌ are comparable to those​ from tubular structure segmentation​‌ in other imaging modalities.​​ Additionally, models trained on​​​‌ resting-state data generalize well​ to images captured during​‌ visual stimulation, highlighting robustness.​​ Although it does not​​ reach the full granularity​​​‌ of ULM, the proposed‌ method provides a practical,‌​‌ non-invasive and cost-effective solution​​ for inferring flow direction–particularly​​​‌ valuable in scenarios where‌ ULM is not available‌​‌ or feasible. Our pipeline​​ shows high linear correlation​​​‌ coefficients between signals from‌ predicted and actual compartments,‌​‌ showcasing its ability to​​ accurately capture blood flow​​​‌ dynamics.

8.6 Methods and‌ frameworks for the analysis‌​‌ of care pathways

Participants:​​ Arnaud Duvermy, Thomas​​​‌ Guyet.

The analysis‌ of care trajectories is‌​‌ essential to assist epidemiologists​​ in their study of​​​‌ the impact of new‌ cares or to stratify‌​‌ patients cohort according to​​ target specific publications for​​​‌ health studies. More specifically,‌ we develop methods and‌​‌ frameworks to cluster patients​​ according to their care​​​‌ trajectoires (represented as sequences‌ of events and as‌​‌ time series).

This year,​​ we had several contributions​​​‌ to this research line:‌

• The major result of‌​‌ the year is the​​ first release of a​​​‌ new platform, nammed TanaT‌ (see Software 7.1.1).‌​‌ TanaT is a Python​​ library designed for advanced​​​‌ temporal sequence analysis, with‌ specialized focus on patient‌​‌ care pathways and complex​​ temporal data structures (trajectories).​​​‌ Its purpose is to‌ gather the methodological developments‌​‌ of the team around​​ the analysis of care​​​‌ trajectories and to provide‌ them to the international‌​‌ public heath community (and​​ more broadly to all​​​‌ team working on timed‌ sequences). A presentation of‌​‌ this new framework to​​ the national 22 and​​​‌ international data science communities‌ 13 initiated the broadcasting‌​‌ of the software in​​ order to make it​​​‌ used by close collaborators‌ and external teams in‌​‌ a short term.
• The​​ difficulty in clustering care​​​‌ pathways, represented by sequences‌ of timestamped events, lies‌​‌ in defining a semantically​​ appropriate dissimilarity and clustering​​​‌ algorithms. In 8,‌ we adapt two methods‌​‌ developed for time series​​ to the clustering of​​​‌ timed sequences: the drop-DTW‌ metric and the DBA‌​‌ approach for the construction​​ of averaged time sequences.​​​‌ These methods are applied‌ in clustering algorithms to‌​‌ propose original and sound​​ clustering algorithms for timed​​​‌ sequences.
• An alternative for‌ the same challenge as‌​‌ been developed in collaboration​​ with the Univerity of​​​‌ Clermont Ferrand to propose‌ an evidential approach of‌​‌ clustering. In 15,​​ 16, we reformulate​​​‌ the Evidential C-Means (ECM)‌ problem for clustering complex‌​‌ data. We propose a​​ new algorithm, Soft-ECM, which​​​‌ consistently positions the centroids‌ of imprecise clusters requiring‌​‌ only a semi-metric. Our​​ experiments show that Soft-ECM​​​‌ present results comparable to‌ conventional fuzzy clustering approaches‌​‌ on numerical data, and​​ we demonstrate its ability​​​‌ to handle mixed data‌ and its benefits when‌​‌ combining fuzzy clustering with​​ semi-metrics such as DTW​​​‌ for time series data.‌ Soft-ECM is also a‌​‌ software that is available​​ for research (see 7.1.2​​​‌).

A part of‌ this work is conducted‌​‌ within the SafePaw project.​​

8.7 Automatic analysis of​​​‌ negation cues and scopes‌ for medical texts in‌​‌ French using language models​​

Participants: Salim Sadoune,​​​‌ Thomas Guyet, Hugues‌ Berry.

Correct automatic‌​‌ analysis of a medical​​​‌ report requires the identification​ of negations and their​‌ scopes. Since most of​​ available training data comes​​​‌ from medical texts in​ English, it usually takes​‌ additional work to apply​​ to non-English languages. In​​​‌ 10, we introduced​ a supervised learning method​‌ for automatically identifying and​​ determining the scopes and​​​‌ negation cues in French​ medical reports using language​‌ models based on BERT.​​ Using a new private​​​‌ corpus of French-language chest​ CT scan reports with​‌ consistent annotation based on​​ clinical data provided by​​​‌ the Radiology Department of​ Lyon University Hospital (Hospices​‌ Civils de Lyon, HCL),​​ we first fine-tuned five​​​‌ available transformer models on​ the negation cue and​‌ scope identification task. Subsequently,​​ we extended the methodology​​​‌ by modifying the optimal​ model to encompass a​‌ wider range of clinical​​ notes and reports (not​​​‌ limited to radiology reports)​ and more heterogeneous annotations.​‌ Lastly, we tested the​​ generated model on its​​​‌ initial mask-filling task to​ ensure there is no​‌ catastrophic forgetting. On a​​ corpus of thoracic CT​​​‌ scan reports annotated by​ four annotators within our​‌ team, our method reaches​​ a F1-score of 99.4%​​​‌ for cue detection and​ 94.5% for scope detection,​‌ thus equaling or improving​​ state-of-the art performance. On​​​‌ more generic biomedical reports,​ annotated with more heterogeneous​‌ rules, the quality of​​ the automatic analysis of​​​‌ course decreases, but our​ best-of-the class model still​‌ delivers very good performance,​​ with F1-scores of 98.2%​​​‌ (cue detection), and 90.9%​ (scope detection). Moreover, we​‌ show that fine-tuning the​​ original model for the​​​‌ negation identification task preserves​ or even improves its​‌ performance on its initial​​ fill-mask task, depending on​​​‌ the lemmatization. Considering the​ performance of our fine-tuned​‌ model for the detection​​ of negation cues and​​​‌ scopes in medical reports​ in French and its​‌ robustness with respect to​​ the diversity of the​​​‌ annotation rules and the​ type of biomedical data,​‌ we conclude that it​​ is suited for use​​​‌ in a real-life clinical​ context.

9.1 Bilateral contracts with​​​‌ industry

Participants: Hugues Berry​, Audrey Denizot,​‌ Thomas Guyet, Jan-Michael​​ Rye, Benjamin Vidal​​​‌, Luc Zimmer,​ Maelle Morange, William​‌ Peoc'H, Andrea Ducos​​, Hana Sebia,​​​‌ Zoe Laffitte, Florian​ Dupeuble, Lucas Perret​‌.

AIstroSight is a​​ joint project-team with the​​​‌ biotech company Theranexus.​ A plain “tutelle” of​‌ the team, Theranexus brings​​ its research expertise in​​​‌ in vitro cell culture,​ disease modelling and imaging,​‌ both in terms of​​ research workforce and data.​​​‌ The stand-alone “convention d'équipe-projet​ commune” of AIstroSight lists​‌ a group of 30​​ rare diseases of the​​​‌ central nervous system, that​ are of direct interest​‌ to Theranexus and thhat​​ are associated with a​​​‌ specific regimen in terms​ of IP and legal​‌ affairs. However, AIstroSight members​​ are allowed to work​​​‌ on pathologies outside this​ list without any restriction​‌ but with a different​​ legal regimen vis-à-vis Theranexus.​​​‌

10.1 International initiatives

10.2 International research visitors​​

10.3 National‌ initiatives

10.4 Regional initiatives​‌

11.1 Promoting scientific​​ activities

11.2​​​‌ Teaching - Supervision -​ Juries - Educational and​‌ pedagogical outreach

11.3​​​‌ Popularization

12.1 Major publications

• 1​‌ articleY.Yulia Dembitskaya​​, C.Charlotte Piette​​​‌, S.Sylvie Perez​, H.Hugues Berry​‌, P. J.Pierre​​ J Magistretti and L.​​Laurent Venance. Lactate​​​‌ supply overtakes glucose when‌ neural computational and cognitive‌​‌ loads scale up.​​Proceedings of the National​​​‌ Academy of Sciences of‌ the United States of‌​‌ America11947November​​ 2022HALDOI
• 2​​​‌ articleA.Audrey Denizot‌, M.María Fernanda‌​‌ Veloz Castillo, P.​​Pavel Puchenkov, C.​​​‌Corrado Calì and E.‌Erik de Schutter.‌​‌ The ultrastructural properties of​​ the endoplasmic reticulum govern​​​‌ microdomain signaling in perisynaptic‌ astrocytic processes.Glia‌​‌October 2025HALDOI​​back to text
• 3​​​‌ inproceedingsA.Andréa Ducos‌, A.Audrey Denizot‌​‌, T.Thomas Guyet​​ and H.Hugues Berry​​​‌. Evaluating PDE discovery‌ methods for multiscale modeling‌​‌ of biological signals..​​Springer LNBI23rd International​​​‌ Conference on Computational Methods‌ in Systems Biology (CMSB)‌​‌15959Lyon, FranceSeptember​​ 2025HAL
• 4 inproceedings​​​‌T.Thomas Guyet and‌ A.Arnaud Duvermy.‌​‌ Towards a Library for​​ the Analysis of Temporal​​​‌ Sequences.AALTD 2025‌ - 10th Workshop on‌​‌ Advanced Analytics and Learning​​ on Temporal DataPorto​​​‌ (Portugal), PortugalSeptember 2025‌HAL
• 5 articleS.‌​‌Salim Sadoune, A.​​Antoine Richard, F.​​​‌François Talbot, T.‌Thomas Guyet, L.‌​‌Loïc Boussel and H.​​Hugues Berry. Automatic​​​‌ analysis of negation cues‌ and scopes for medical‌​‌ texts in French using​​ language models.Computers​​​‌ in Biology and Medicine‌1972025, 110795‌​‌HALDOI
• 6 article​​H.Hana Sebia,​​​‌ T.Thomas Guyet,‌ M.Mickaël Pereira,‌​‌ M.Marco Valdebenito,​​ H.Hugues Berry and​​​‌ B.Benjamin Vidal.‌ Vascular Segmentation of Functional‌​‌ Ultrasound Images using Deep​​ Learning.Computers in​​​‌ Biology and Medicine194‌May 2025, 110377‌​‌HALDOI

12.2 Publications​​ of the year

12.3 Cited​​ publications

• 26 articleS.​​​‌Shofiul Azam, M.‌Md Haque, M.‌​‌Md Jakaria, S.-H.​​Song-Hee Jo, I.-S.​​​‌In-Su Kim and D.-K.‌Dong-Kug Choi. G-Protein-Coupled‌​‌ Receptors in CNS: A​​ Potential Therapeutic Target for​​​‌ Intervention in Neurodegenerative Disorders‌ and Associated Cognitive Deficits‌​‌.Cells92​​2020, 506back​​​‌ to text
• 27 inproceedings‌A.Ana\"is Badoual,‌​‌ M.Misa Arizono,​​ A.Audrey Denizot,​​​‌ M.Mathieu Ducros,‌ H.Hugues Berry,‌​‌ U.U Valentin Nägerl​​ and C.Charles Kervrann​​​‌. Simulation of Astrocytic‌ Calcium Dynamics in Lattice‌​‌ Light Sheet Microscopy Images​​.IEEE International Symposium​​​‌ on Biomedical Imaging, ISBI‌2021, 135-139back‌​‌ to text
• 28 article​​N.N. Bazargani and​​​‌ A.A. Attwel.‌ Astrocyte calcium signaling: the‌​‌ third wave.Nature​​ Neuroscience192016,​​​‌ 182-189back to text‌
• 29 articleA.Anton‌​‌ Bespalov, T.Thomas​​ Steckler, B.Bruce​​​‌ Altevogt, E.Elena‌ Koustova, P.Phil‌​‌ Skolnick, D.Daniel​​ Deaver, M. J.​​​‌Mark J. Millan,‌ J. F.Jesper F.‌​‌ Bastlund, D.Dario​​ Doller, J.Jeffrey​​​‌ Witkin, P.Paul‌ Moser, P.Patricio‌​‌ O'Donnell, U.Ulrich​​ Ebert, M. A.​​​‌Mark A. Geyer,‌ E.Eric Prinssen,‌​‌ T.Theresa Ballard and​​ M.Malcolm Macleod.​​​‌ Failed trials for central‌ nervous system disorders do‌​‌ not necessarily invalidate preclinical​​ models and drug targets​​​‌.Nature Reviews Drug‌ Discovery152016,‌​‌ 516back to text​​
• 30 inproceedingsT.T.​​​‌ Bie. Subjective interestingness‌ in exploratory data mining‌​‌.International Symposium on​​ Intelligent Data Analysis2013​​​‌, 19-31back to‌ text
• 31 inproceedingsM.‌​‌M. Bienvenu. Ontology-mediated​​ query answering: harnessing knowledge​​​‌ to get more from‌ data.Proceedings IJCAI'16‌​‌2016, 4058-4061back​​​‌ to text
• 32 article​S.S.L. Brunton,​‌ J.J.L. Proctor and​​ J.J.N. Kutz.​​​‌ Discovering governing equations from​ data by sparse identification​‌ of nonlinear dynamical systems​​.Proc Natl Acad​​​‌ Sci USA1132016​, 3932back to​‌ text
• 33 articleA.​​A. Callahan. ACE:​​​‌ the Advanced Cohort Engine​ for searching longitudinal patient​‌ records.J Am​​ Med Inform Assoc28​​​‌72021, 1468-1479​back to text
• 34​‌ incollectionE.E.W.M. Chin​​ and E.E.L.K. Goh​​​‌. MeCP2 Dysfunction in​ Rett Syndrome and Neuropsychiatric​‌ Disorders.Psychiatric Disorders:​​ Methods and Protocols, Methods​​​‌ in Molecular Biology2011​Humana, New York, NY​‌2019, 573-592back​​ to text
• 35 article​​​‌A.Antony Cougnoux,​ J.Julia Yerger,​‌ M.Mason Fellmeth,​​ J.Jenny Serra Vinardell​​​‌, K.Kyle Martin​, F.Fatemeh Navid​‌, J.James Iben​​, C.Christopher Wassif​​​‌, N.Niamh Cawley​ and F.Forbe Porter​‌. Single Cell Transcriptome​​ Analysis of Niemann-Pick Disease,​​​‌ Type C1 Cerebella.​Int J Mol Sci​‌212020, 5368​​back to text
• 36​​​‌ articleY.Yihui Cui​, V.Vincent Paille​‌, H.Hao Xu​​, S.Stéphane Genet​​​‌, B.Bruno Delord​, E.Elodie Fino​‌, H.Hugues Berry​​ and L.Laurent Venance​​​‌. Endocannabinoids mediate bidirectional​ striatal spike-timing dependent plasticity​‌.J Physiol593​​132015, 2833-2849​​​‌back to text
• 37​ articleY.Yihui Cui​‌, I.Ilya Prokin​​, A.Alexandre Mendes​​​‌, H.Hugues Berry​ and L.Laurent Venance​‌. Robustness of STDP​​ to spike timing jitter​​​‌.Scientific Reports8​2018, 8139back​‌ to text
• 38 article​​Y.Yihui Cui,​​​‌ I.Ilya Prokin,​ H.Hao Xu,​‌ B.Bruno Delord,​​ S.Stephane Genet,​​​‌ L.Laurent Venance and​ H.Hugues Berry.​‌ Endocannabinoid dynamics gate spike-timing​​ dependent depression and potentiation​​​‌.eLife52016​, e13185back to​‌ text
• 39 articleY.​​Yihui Cui, Y.​​​‌Yan Yang, Z.​Zheyi Ni, Y.​‌Yiyan Dong, G.-H.​​Guo-Hong Cai, A.​​​‌Alexandre Foncelle, S.​Shuangshuang Ma, K.​‌Kangning Sang, S.​​Siyang Tang, Y.​​​‌Yuezhou Li, Y.​Ying Shen, H.​‌Hugues Berry, S.-X.​​Sheng-Xi Wu and H.​​​‌Hailan Hu. Astroglial-Kir4.1​ in lateral habenula drives​‌ neuronal bursts in depression​​.Nature5542018​​​‌, 323-327back to​ text
• 40 articleA.​‌Audrey Denizot, M.​​Misa Arizono, V.​​​‌ U.Valentin U Nagerl​, H.Hedi Soula​‌ and H.Hugues Berry​​. Simulation of calcium​​​‌ signaling in fine astrocytic​ processes: effect of spatial​‌ properties on spontaneous activity​​.PLoS Comput Biol​​​‌1582019,​ 1006795back to text​‌
• 41 articleA.Audrey​​ Denizot, C.Corrado​​​‌ Cali, H.Hugues​ Berry and E.Erik​‌ De Schutter. Stochastic​​ spatially-extended simulations predict the​​​‌ effect of ER distribution​ on astrocytic microdomain Ca2+​‌ activity.ACM NanoCom​​202021, 1-5​​back to text
• 42​​​‌ articleS.S. Erikainen‌ and S.S. Chan‌​‌. Contested futures: envisioning​​ "Personalized', Stratified," and "Precision"​​​‌ medicine.New Genet‌ Soc3832019‌​‌, 308-330back to​​ text
• 43 articleT.​​​‌Tracey Filzen, P.‌Peter Kutchukian, J.‌​‌Jeffrey Hermes, J.​​Jing Li and M.​​​‌Matthew Tudor. Representing‌ high throughput expression profiles‌​‌ via perturbation barcodes reveals​​ compound targets.PLoS​​​‌ Comput Biol132‌2017, 1005335back‌​‌ to text
• 44 article​​S.S. Finkbeiner.​​​‌ Bridging the Valley of‌ Death of therapeutics for‌​‌ neurodegeneration.Nature Medicine​​162010, 1227-1232​​​‌back to text
• 45‌ articleA.Alexandre Foncelle‌​‌, A.Alexandre Mendes​​, J.Joanna Jedrzejewska-Szmek​​​‌, S.Silvana Valtcheva‌, H.Hugues Berry‌​‌, K.Kim Blackwell​​ and L.Laurent Venance​​​‌. Modulation of spike-timing‌ dependent plasticity: towards the‌​‌ inclusion of a third​​ factor in computational models​​​‌.Frontiers in Computational‌ Neuroscience122018,‌​‌ 49back to text​​
• 46 articleJ.Jessica​​​‌ Freitas, K.Kipp‌ Johnson, E.Eddye‌​‌ Golden, G.Girish​​ Nadkarni, E.Erwin​​​‌ Bottinger, B.Benjamin‌ Glicksberg and R.Riccardo‌​‌ Miotto. Phe2vec: Automated​​ disease phenotyping based on​​​‌ unsupervised embeddings from electronic‌ health records.Patterns‌​‌292021,​​ 100337back to text​​​‌
• 47 articleK.Karl‌ Friston, A.Andrea‌​‌ Mechelli, R.Robert​​ Turner and C.C.J.​​​‌ Price. Nonlinear Responses‌ in fMRI: The Balloon‌​‌ Model, Volterra Kernels, and​​ Other Hemodynamics.NeuroImage​​​‌122000, 466-477‌back to text
• 48‌​‌ articleN. J.Nao​​ J. Gamo, M.​​​‌ R.Michelle R. Birknow‌, D.Danielle Sullivan‌​‌, M. A.Mari​​ A. Kondo, Y.​​​‌Yasue Horiuchi, T.‌Takeshi Sakurai, B.‌​‌ S.Barbara S. Slusher​​ and A.Akira Sawa​​​‌. Valley of death:‌ A proposal to build‌​‌ a "translational bridge" for​​ the next generation.​​​‌Neurosci Res1152017‌, 1-4back to‌​‌ text
• 49 articleG.​​Giuseppe Gangarossa, S.​​​‌Sylvie Perez, Y.‌Yulia Dembitskaya, I.‌​‌Ilya Prokin, H.​​Hugues Berry and L.​​​‌Laurent Venance. BDNF‌ controls bidirectional endocannabinoid-plasticity at‌​‌ corticostriatal synapses.Cerebral​​ Cortex1302020​​​‌, 197-217back to‌ text
• 50 articleR.‌​‌Rohan Gupta, D.​​Devesh Srivastava, M.​​​‌Mehar Sahu, S.‌Swati Tiwari, R.‌​‌Rashmi Ambasta and P.​​Pravir Kumar. Artificial​​​‌ intelligence to deep learning:‌ machine intelligence approach for‌​‌ drug discovery.Molecular​​ Diversity252021,​​​‌ 1315-1360back to text‌
• 51 articleS.Sima‌​‌ Hodavance, C.Clarice​​ Gareri, R.Rachel​​​‌ Torok and H.Howard‌ Rockman. G Protein-coupled‌​‌ Receptor Biased Agonism.​​J Cardiovasc Pharmacol67​​​‌32016, 193-202‌back to text
• 52‌​‌ articleJ.J.P.K. Ip​​, N.N. Mellios​​​‌ and M.M. Sur‌. Rett syndrome: insights‌​‌ into genetic, molecular and​​ circuit mechanisms.Nat​​​‌ Rev Neurosci192018‌, 368-382back to‌​‌ textback to text​​​‌
• 53 articleR.Renaud​ Jolivet, J.Jay​‌ Coggan, I.Igor​​ Allaman and P.Pierre​​​‌ Magistretti. Multi-timescale Modeling​ of Activity-Dependent Metabolic Coupling​‌ in the Neuron-Glia-Vasculature Ensemble​​.PLoS Comput Biol​​​‌112015, 1004036​back to text
• 54​‌ articleM.Matthew Kiernan​​, S.Steve Vucic​​​‌, K.Kevin Talbot​, C.Christopher McDermott​‌, O.Orla Hardiman​​, J.Jeremy Shefner​​​‌, A.Ammar Al-Chalabi​, W.William Huynh​‌, M.Merit Cudkowicz​​, P.Paul Talman​​​‌, L.Leonard Berg​, T.Thanuja Dharmadasa​‌, P.Paul Wicks​​, C.Claire Reilly​​​‌ and M.Martin Turner​. Improving clinical trial​‌ outcomes in amyotrophic lateral​​ sclerosis.Nature Reviews​​​‌ Neurology1712 2020​back to text
• 55​‌ articleC.Charlotte Kloft​​, M.Miriam Trame​​​‌ and C.Christoph Ritter​. Systems pharmacology in​‌ drug development and therapeutic​​ use — A forthcoming​​​‌ paradigm shift.Eur​ J Pharm Sci94​‌2016, 4-14back​​ to text
• 56 article​​​‌N.N. Lim and​ P.P. Pavlidis.​‌ Evaluation of connectivity map​​ shows limited reproducibility in​​​‌ drug repositioning.Sci​ Rep1112013​‌, 17624back to​​ text
• 57 articleB.​​​‌Balázs Lőrincz, E.​Elizabeth Jury, M.​‌Michal Vrablik, M.​​Murali Ramanathan and T.​​​‌Tomas Uher. The​ role of cholesterol metabolism​‌ in multiple sclerosis: From​​ molecular pathophysiology to radiological​​​‌ and clinical disease activity​.Autoimmunity Rev21​‌62022, 103088​​back to text
• 58​​​‌ incollectionP.Pierre Monnin​, C.Chedy Raïssi​‌, A.Amedeo Napoli​​ and A.Adrien Coulet​​​‌. Discovering alignment relations​ with Graph Convolutional Networks:​‌ a biomedical case study​​.2021, 03452182​​​‌back to text
• 59​ articleA.A. Newman-Tancredi​‌. Translating biased agonists​​ from molecules to medications:​​​‌ Serotonin 5-HT1A receptor functional​ selectivity for CNS disorders​‌.Pharmacology Therapeutics107937​​2021back to text​​​‌
• 60 articleP.Patricio​ O'Donnell, L. B.​‌Laura B. Rosen,​​ R.Robert Alexander,​​​‌ V. N.Venkatesh N.​ Murthy, C. H.​‌Ceri H. Davies and​​ E.Emiliangelo Ratti.​​​‌ Strategies to Address Challenges​ in Neuroscience Drug Discovery​‌ and Development.International​​ Journal of Neuropsychopharmacology22​​​‌2019, 445 -​ 448back to text​‌back to text
• 61​​ articleV.V.I. Pérez-Nueno​​​‌. Using quantitative systems​ pharmacology for novel drug​‌ discovery.Expert Opin​​ Drug Discov1012​​​‌2015, 1315-1331back​ to text
• 62 article​‌S.S.R. Pfeffer.​​ NPC intracellular cholesterol transporter​​​‌ 1 (NPC1)-mediated cholesterol export​ from lysosomes.J​‌ Biol Chem2945​​2019, 1706-1709back​​​‌ to textback to​ text
• 63 inproceedingsZ.​‌Z. Qian, K.​​K. Kacprzyk and M.​​​‌M. Schaar. D-code:​ discovering closed-form ODEs from​‌ observed trajectories.Proceedings​​ ICLR 20212021back​​​‌ to text
• 64 article​M.M. Raissi,​‌ P.P. Perdikaris and​​ G.G.E. Karniadakis.​​​‌ Physics-informed neural networks :​ A deep-learning framework for​‌ solving forward and inverse​​ problem involving nonlinear partial​​ differential equations.J​​​‌ Comput Phys3782019‌, 686-707back to‌​‌ text
• 65 articleC.​​Chaitanya Shivade, P.​​​‌Preethi Raghavan, E.‌Eric Fosler-Lussier, P.‌​‌Peter Embi, N.​​Noemie Elhadad, S.​​​‌Stephen Johnson and A.‌Albert Lai. A‌​‌ review of approaches to​​ identifying patient phenotype cohorts​​​‌ using electronic health records‌.J Am Med‌​‌ Inform Assoc212​​2014, 221-230back​​​‌ to text
• 66 article‌S.Stephen Shovlin and‌​‌ D.Daniela Tropea.​​ Transcriptome level analysis in​​​‌ Rett syndrome using human‌ samples from different tissues‌​‌.Orphanet J Rare​​ Dis1312018​​​‌, 113back to‌ text
• 67 inproceedingsA.‌​‌A. Silberschatz and A.​​A. Tuzhilin. On​​​‌ subjective measures of interestingness‌ in knowledge discovery.‌​‌Proceedings KDD'951995,​​ 275-281back to text​​​‌
• 68 articleT.Tomonori‌ Takeuchi, A.Adrian‌​‌ Duszkiewicz and R.Richard​​ Morris. The synaptic​​​‌ plasticity and memory hypothesis:‌ encoding, storage and persistence‌​‌.Philos Trans R​​ Soc Lond B369​​​‌2014, 20130288back‌ to text
• 69 article‌​‌J.Jessica Vamathevan,​​ D.Dominic Clark,​​​‌ P.Paul Czodrowski,‌ I.Ian Dunham,‌​‌ E.Edgardo Ferran,​​ G.George Lee,​​​‌ B.Bin Li,‌ A.Anant Madabhushi,‌​‌ P.Parantu Shah,​​ M.Michaela Spitzer and​​​‌ S.Shanrong Zhao.‌ Applications of machine learning‌​‌ in drug discovery and​​ development.Nature Reviews​​​‌ Drug Discovery182019‌, 463-477back to‌​‌ text
• 70 articleM.​​M.T. Vanier. Niemann-Pick​​​‌ disease type C.‌Orphanet J Rare Dis‌​‌5162010back​​ to text
• 71 article​​​‌A.A. Verkhratsky and‌ M.M. Nedergaard.‌​‌ Physiology of Astroglia.​​Physiol Rev981​​​‌2018, 239-289back‌ to text
• 72 article‌​‌B.Benjamin Vidal,​​ S.Sylvain Fieux,​​​‌ J.Jérôme Redouté,‌ M.Marjorie Villien,‌​‌ F.Frédéric Bonnefoi,​​ D.Didier Bars,​​​‌ A.Adrian Newman-Tancredi,‌ N.Nicolas Costes and‌​‌ L.Luc Zimmer.​​ In vivo biased agonism​​​‌ at 5-HT1A receptors: characterisation‌ by simultaneous PET/MR imaging‌​‌.Neuropsychopharmacology2018,​​ 2310-2319back to text​​​‌
• 73 articleR.R.‌ Weinshilboum. Inheritance and‌​‌ Drug Response.N.​​ Engl J Med348​​​‌2003, 529-537back‌ to text
• 74 article‌​‌H.H. Xu,​​ S.S. Perez,​​​‌ A.A. Cornil,‌ B.B. Detraux,‌​‌ I.I. Prokin,​​ Y.Y. Cui,​​​‌ B.B. Degos,‌ H.H. Berry,‌​‌ A.A. De Kerchove​​ D'exaerde and L.L.​​​‌ Venance. Dopamine- endocannabinoid‌ interactions mediate spike-timing dependent‌​‌ potentiation in the striatum​​.Nature Communications9​​​‌2018, 4118back‌ to text
• 75 inproceedings‌​‌K.Kejing Yin,​​ D.Dong Qian,​​​‌ K.-W.Kwok-Wai Cheung,‌ B.Benjamin Fung and‌​‌ J.Jonathan Poon.​​ Learning phenotypes and dynamic​​​‌ patient representations via rnn‌ regularized collective non-negative tensor‌​‌ factorization.Proceedings of​​ the 2019 AAAI Conference​​​‌ on Artificial Intelligence2019‌, 1246-1253back to‌​‌ text
• 76 articleM.​​​‌Min Zhang, D.​Diana Strnatka, C.​‌Carolyn Donohue, J.​​Janice Hallows, I.​​​‌Inez Vincent and R.​Robert Erickson. Astrocyte-only​‌ Npc1 reduces neuronal cholesterol​​ and triples life span​​​‌ of Npc1-/- mice.​J Neurosci Res86​‌2008, 2848-2856back​​ to text