Synthesis and simulation of formal models of biological systems
19th July 2024
At Sémidoc, LaBRI, Bordeaux.
Slides
👋 Hej!
- 💼 Since september 2024, I am a "maîtresse de conférence" (≈ associate professor) at the university of Caen, Normandy. I teach computer science in the "BUT informatique", and I do research in the team CODAG of the GREYC.
- 🧑🎓 Before that, I was a postdoctoral researcher at LaBRI (team MTV) under the direction of Loïc Paulevé. I did my PhD in CRAN and LORIA (capsid team), under the supervision of Taha Boukhobza & Malika Smail-Tabbone (check my thesis here).
- 💻 I describe my research interest in a dedicated section (here). TLDR : I work on the formal analysis of Boolean networks and reaction networks. Check out my publications;
- 🌍 I currently live in Caen (🇫🇷), previously I was in Bordeaux (🇫🇷), Iowa City (🇺🇸), Nancy (🇫🇷), Lille (🇫🇷), and Paris (🇫🇷).
-
🤹 Also, I am :
- 🐍 a Python lover (the programming language), also fond of Answer-Set Programming (ASP), but not much of R;
- 🐀 a rat care-taker (the cute animal);
- 🎸🎺 a musician (guitar and trumpet).
You may want:
- 📧 my e-mail address: athenais.vaginay🐌unicaen.fr;
- 📰 my academic papers
on HAL
/ Google scholar
/ dblp
(/!\ dblp covers publications from computer science only) ; -
my LinkedIn (updated in August 2023);
-
my GitHub,
but note that the code of my research projects spreads
on several lab-specific GitLab instances
that are accessible on request.
- 📭 my work address (I love letters!):
lab: Bâtiment Sciences 3 (GREYC)
6 Boulevard du Maréchal Juin
CS 14032, 14032 CAEN cedex 5
teaching: IUT de Caen Département Info
3 rue Anton Tchekhov
14123 Ifs
Feel free to get in touch with me (in french or english). :)
Research
My field is those of formal systems biology. During my Ph.D., I proposed a method to convert a reaction network to a (set of) compatible Boolean automata networks. The manuscript is here.
What I want is ultimately to (1) participate in defining the formal correspondence between those two widely used formalisms, (2) propose new alternatives for the analysis of reaction networks (3) improve the Boolean networks synthesis methods, when applied on real biological data. Towards those goals, I define my problems in terms of satisfaction of logical constraints, I use formal tools such as static analysis and abstract interpretation to define the constraints, and SAT solvers and ASP (clingo) for my implementations to beat the curse of complexity.
The list of my publications is below.
Looking for...
💬 Talks
The list below should be more or less exhaustive. The slides are available for most of the presentations.List
Constraint-based abstraction of reaction networks to Boolean networks
26th Feb 2024
At CLC seminar, Computer Science department of the university of Iowa, Iowa City.
Slides
My presentation delves into the pivotal role of formal methods during the two steps of the conversion process: (1) in the definition of the properties of the input reaction network we want to preserve and (2) the synthesis of Boolean networks complying with these properties. The evaluation of SBML2BNET on synthetic examples and on real-world reaction networks from the repository BioModels has demonstrated its effectiveness. The perspectives mainly concern the practical relevance of the Boolean networks we synthesize, as well as the formal exploration of the relationship with other semantics of reaction networks.
Abstract
Reaction networks and Boolean networks are two prominent approaches for modeling biological systems. They rely on drastically different philosophies: the former consists of a detailed description with continuous time and values, and the latter uses discrete steps and Boolean values. The automatic conversion of reaction networks to Boolean networks is an important challenge of symbolic systems biology, which I address with the SBML2BNET pipeline.My presentation delves into the pivotal role of formal methods during the two steps of the conversion process: (1) in the definition of the properties of the input reaction network we want to preserve and (2) the synthesis of Boolean networks complying with these properties. The evaluation of SBML2BNET on synthetic examples and on real-world reaction networks from the repository BioModels has demonstrated its effectiveness. The perspectives mainly concern the practical relevance of the Boolean networks we synthesize, as well as the formal exploration of the relationship with other semantics of reaction networks.
Constraint-based abstraction of reaction networks to Boolean networks
5th December 2023
At GT CODAG, GREYC, Caen, France.
Slides
Abstract
Reaction networks and Boolean networks are two prominent approaches for modeling biological systems. They rely on drastically different philosophies : the former consists of a detailed process-centred description with continuous time and values and the latter of an abstract species-centred description with discrete steps and Boolean values. In order to better understand the relationship between these two formalisms, my PhD work focused on converting reaction networks to Boolean networks. This presentation delves into the pivotal role of constraints during the two steps of the process: (1) in the definition of the properties of the input reaction network we want to preserve and (2) the synthesis of the Boolean networks complying with these properties. So far, the evaluation of the approach on toy examples and real-world reaction networks from the repository BioModels, has demonstrated its effectiveness. The perspectives mainly concern the practical relevance of the Boolean networks we synthesise, as well as the formal exploration of the relationship with other semantics of reaction networks.From reaction networks to Boolean networks: why and how
14 November 2023
At CANA seminar, LIS, Marseille, France.
Slides
Abstract
One way to get new insights about complex biological systems is to convert between modelling formalisms. Here, we deal with the conversion of reaction networks interpreted with the differential semantics, into Boolean networks. The conversion is particularly challenging, as it requires a drastic change in perspective: from a process-centred description with continuous time and values to a species-centred description with discrete steps and Boolean values. The conversion I present here is based on my PhD work. It aims at preserving properties from the input reaction network, such as its structure (the direct influences between the components) and its binarised transient dynamics (the transitions between the Boolean configurations). We will see how to extract the structure and dynamics of a reaction network, and how to use answer-set programming synthesise complying Boolean networks. So far, the evaluation of the approach on toy examples and real-world reaction networks from the repository BioModels, has demonstrated it effectiveness in synthesising Boolean networks complying with the input reaction networks. It also paved the way of the formal study of the relationship between the differential semantics of reaction network and Boolean networks. The perspectives mainly concern the practical relevance of the Boolean networks we synthesise, as well as the formal exploration of the relationship with other semantics of reaction networks.Synthesis of Boolean Networks from the Structure and Dynamics of Reaction Networks
7th July 2023
At Loria, Nancy, France (PhD defense).
Synthesis of Boolean Networks from the Structure and Dynamics of Reaction Networks
3rd July 2023
At Thematic workshop on networks and biological models inference, CIRM, Marseille, France (invited presentation).
Sélection et analyse de modèles pour les réseaux biologiques
Systems biology, reaction and Boolean networks: introduction and conversion from reaction network to Boolean networks
Differential Equations and differential invariants
22th August 2022
At Vino (summer school / reading group), Campo Tures, Italy (book chapter presentation).
ASKeD-BN: Automatic Synthesis of Boolean Networks from Knowledge and Data
From Chemical Reaction Networks to Boolean Networks, Automatically
From Quantitative SBML to Boolean Network
1st December 2021
Modelling Biological Systems with Boolean Networks
SBML2BN: an Integrative Pipeline for the Synthesis of Boolean Networks from SBML Models
ASKeD-BN: Automatic Synthesis of Boolean Networks from Knowledge and Data
ASKeD-BN: Automatic Synthesis of Boolean Networks from Knowledge and Data
Towards an automatic conversion from SBML core to SBML qual
Quick Presentation
caspid "tea-time" seminars
Those are monthly scheduled seminars of the capsid team During my PhD, the talks below have mostly been used to keep my team updated about my work, and to rehearse presentations for conferences and other seminars. This is why I keep the list separated from the other talks.List
ASKeD-BN: Automatic Synthesis of Boolean Networks from Knowledge and Data (OLA rehearsal)
Jun. 2021
caspid "tea-time" seminar, Nancy, France
Constrained Enumeration of Boolean Networks from Biological Data and Knowledge (prepare OLA submission)
The Search Space of the Logical Function Synthesis Problem — Application for Biological Systems
Automatic Transformation from Reaction Models to Boolean Models Using Answer-Sets Constrained by a Topology and an Abstracted Dynamic
What I've learn this summer: recap Bioregul summer school (Porquerolle) and JeBIF + JOBIM (Nantes)
30 Jul. 2019
caspid "tea-time" seminar, Nancy, France
Introduction to Critical Thinking (in french)
Presentation of my Thesis Project
📄 Publications
I put the pdf of my papers on HAL.List
This is the exhaustive list of peer-reviewed publications I co-authored. Color code: Journal Conference Thesis
- S. Chevalier, D. Boyenval, G. Magaña-López, T. Roncalli, A. Vaginay, L. Paulevé BoNesis: a Python-based declarative environment for the verification, reprogramming, and synthesis of Most Permissive Boolean networks. CMSB 2024: 22nd International Conference on Computational Methods in Systems Biology, 2024, Pisa, Italy.
- J. Niehren, C. Lhoussaine, and A. Vaginay, Core SBML and its formal semantics, in Computational methods in systems biology - 21st international conference, CMSB 2023, luxembourg city, luxembourg, september 13-15, 2023, proceedings, J. Pang and J. Niehren, Eds., in Lecture notes in computer science, vol. 14137. Springer, 2023, pp. 124–143. doi: 10.1007/978-3-031-42697-1_9.
- A. Vaginay, Synthesis of Boolean networks from the structure and dynamics of reaction networks. phd thesis, University of Lorraine, Nancy, France, 2023. https://tel.archives-ouvertes.fr/tel-04257373
- A. Vaginay, T. Boukhobza, and M. Smaïl-Tabbone, From quantitative SBML models to Boolean networks, Appl. Netw. Sci., vol. 7, no. 1, p. 73, 2022, doi: 10.1007/S41109-022-00505-8.
- Hirtz A, Lebourdais N, Rech F, Bailly Y, Vaginay A, Smaïl-Tabbone M, Dubois-Pot-Schneider H, Dumond H. GPER Agonist G-1 Disrupts Tubulin Dynamics and Potentiates Temozolomide to Impair Glioblastoma Cell Proliferation. Cells. 2021 Dec 7; 10(12):3438. doi: 10.3390/cells10123438.
- A. Vaginay, T. Boukhobza, and M. Smaïl-Tabbone, From quantitative SBML models to boolean networks, in Complex networks & their applications X - volume 2, proceedings of the tenth international conference on complex networks and their applications COMPLEX NETWORKS 2021, madrid, spain, november 30 - december 2, 2021, R. M. Benito, C. Cherifi, H. Cherifi, E. Moro, L. M. Rocha, and M. Sales-Pardo, Eds., in Studies in computational intelligence, vol. 1016. Springer, 2021, pp. 676–687. doi: 10.1007/978-3-030-93413-2_56.
- J. Niehren, A. Vaginay, and C. Versari, Abstract simulation of reaction networks via boolean networks, in Computational methods in systems biology - 20th international conference, CMSB 2022, bucharest, romania, september 14-16, 2022, proceedings, I. Petre and A. Paun, Eds., in Lecture notes in computer science, vol. 13447. Springer, 2022, pp. 21–40. doi: 10.1007/978-3-031-15034-0_2.
- A. Vaginay, T. Boukhobza, and M. Smaïl-Tabbone, Automatic synthesis of boolean networks from biological knowledge and data, in Optimization and learning - 4th international conference, OLA 2021, catania, italy, june 21-23, 2021, proceedings, B. Dorronsoro, L. Amodeo, M. Pavone, and P. Ruiz, Eds., in Communications in computer and information science, vol. 1443. Springer, 2021, pp. 156–170. doi: 10.1007/978-3-030-85672-4_12.
🧑🏫 Teaching
The class links items are for the students to whom I give classes (for most, they require connection on the uni website), but do not hesitate to reach out if you want to know more about it.List
- 2024-2025 (BUT info) Page dédiée
- 2021-2022 (IDMC - L1 MIASH -- Anne Boyer)
Algorithmique et structure de données (semestre 1)
Arche - 2021-2022 (IDMC - L1 MIASH -- Laurent Vigneron)
Programmation C (semestre 1) - 2021-2022 (IDMC - L1 MIASH -- Armelle Brun)
Programmation C (semestre 2)
Arche - 2021-2022 (IDMC - L2 MIASH -- Maxime Amblard)
Traitement automatique des langues (semestre 2)
Arche - 2021-2022 (IDMC - M1 SC, SDL, TAL -- Mathieu d'Aquin)
Programmation Python (semestre 1)
Arche - 2021-2022 (IDMC - M2 TAL et SC)
Méthodologie, LaTeX et gestion de bibliographie
Arche - 2021-2022 (IDMC - M1 TAL et SC)
LaTeX et gestion de bibliographie
Arche - 2019-2020 (IUT Brabois - DUT GBS 1A -- Alexandre Desforges)
Bureautique et recherche documentaire -- Outils informatiques (semestre 1)
Arche - 2019-2020 (IUT Brabois - DUT GB2A 1A -- Damien Cornu)
Bureautique -- Informatique-Word-Powerpoint (semestre 1)
Arche - 2016-2017 (Univ. Paris Diderot - L1 EA)
Bureautique, vie numérique et recherche documentaire -- Outils pour la Bureautique et internet (semestre 2)
Moodle - 2014–2015 (Univ. Paris Diderot - tutorat L1)
Bureautique, vie numérique et recherche documentaire -- Outils pour la Bureautique et internet