A5035679840- Neuroblastoma Research and Treatments
- Gene expression and cancer classification
- Statistical Methods and Inference
- Genomics and Chromatin Dynamics
- Machine Learning and Data Classification
- Neural Networks and Applications
- Soil Carbon and Nitrogen Dynamics
- Cancer, Hypoxia, and Metabolism
- ECG Monitoring and Analysis
- Genomics and Phylogenetic Studies
- EEG and Brain-Computer Interfaces
- Cancer Genomics and Diagnostics
- Bayesian Modeling and Causal Inference
- Heart Rate Variability and Autonomic Control
- Data Analysis with R
- Genomic variations and chromosomal abnormalities
- Genetic Associations and Epidemiology
- Data Mining Algorithms and Applications
- Imbalanced Data Classification Techniques
- Bioinformatics and Genomic Networks
- Molecular Biology Techniques and Applications
- Time Series Analysis and Forecasting
- Data Visualization and Analytics
- Machine Learning and Algorithms
- Algorithms and Data Compression
Université de Sherbrooke
2025
Northern Arizona University
2018-2024
Institut des Sciences des Plantes de Paris Saclay
2021
McGill University and Génome Québec Innovation Centre
2014-2019
McGill University
2014-2018
Institut Curie
2014-2017
Université Paris Sciences et Lettres
2016
Inserm
2014-2016
Tokyo Institute of Technology
2014
École Nationale Supérieure des Mines de Paris
2014
Many common approaches to detecting changepoints, for example based on statistical criteria such as penalised likelihood or minimum description length, can be formulated in terms of minimising a cost over segmentations. We focus class dynamic programming algorithms that solve the resulting minimisation problem exactly, and thus find optimal segmentation under given criteria. The standard implementation these methods have computational scales at least quadratically length time-series....
The tumor genomic copy number profile is of prognostic significance in neuroblastoma patients. We have studied the cell-free DNA (cfDNA) and compared this with primary arrayCGH (aCGH) at diagnosis.In 70 patients, cfDNA profiling was performed using OncoScan platform. profiles were classified according to overall pattern, including numerical chromosome alterations (NCA), segmental (SCA), MYCN amplification (MNA).Interpretable dynamic obtained 66 52 cases, respectively. An identical between...
Neuroblastoma is characterized by substantial clinical heterogeneity. Despite intensive treatment, the survival rates of high-risk neuroblastoma patients are still disappointingly low. Somatic chromosomal copy number aberrations have been shown to be associated with patient outcome, particularly in low- and intermediate-risk patients. To improve outcome prediction neuroblastoma, we aimed design a prognostic classification method based on aberrations.In an international collaboration,...
Survival regression is used to estimate the relation between time-to-event and feature variables, important in application domains such as medicine, marketing, risk management, sales management. Nonlinear tree based machine learning algorithms implemented libraries XGBoost, scikit-learn, LightGBM, CatBoost are often more accurate practice than linear models. However, existing state-of-the-art implementations of tree-based models have offered limited support for survival regression. In this...
Chatbots are often designed to mimic social roles attributed humans. However, little is known about the impact of using language that fails conform associated role. Our research draws on sociolinguistic investigate how a chatbot’s choices can adhere expected role agent performs within context. We seek understand whether chatbots design should account for linguistic register. This analyzes register differences play in shaping user’s perception human-chatbot interaction. produced parallel...
Summary Calcium imaging data promises to transform the field of neuroscience by making it possible record from large populations neurons simultaneously. However, determining exact moment in time at which a neuron spikes, calcium set, amounts non-trivial deconvolution problem is critical importance for downstream analyses. While number formulations have been proposed this task recent literature, article, we focus on formulation recently Jewell and Witten (2018. Exact spike train inference via...
Microorganisms are found in almost every environment, including soil, water, air and inside other organisms, such as animals plants. While some microorganisms cause diseases, most of them help biological processes decomposition, fermentation nutrient cycling. Much research has been conducted on the study microbial communities various environments how their interactions relationships can provide insight into diseases. Co-occurrence network inference algorithms us understand complex...
Precise quantification of greenhouse gas (GHG) emissions is important for better urban sustainability. Transportation one the primary contributing sources emissions. To quantify on-road GHG emissions, it essential to decode fleet distribution. However, globally, many cities do not have infrastructure calculate a Therefore, there will always be an uncertain error in estimation. very high-resolution satellite data can helpful overcome this gap due its global temporal coverage. Hence, study...
Many models have been proposed to detect copy number alterations in chromosomal profiles, but it is usually not obvious decide which most effective for a given data set. Furthermore, methods smoothing parameter that determines the of breakpoints and must be chosen using various heuristics. We present three contributions profile model selection. First, we propose select degree smoothness maximizes agreement with visual breakpoint region annotations. Second, develop cross-validation procedures...
Many peak detection algorithms have been proposed for ChIP-seq data analysis, but it is not obvious which algorithm and what parameters are optimal any given dataset. In contrast, regions with without peaks can be easily labeled by visual inspection of aligned read counts in a genome browser. We propose supervised machine learning approach using labels that encode qualitative judgments about genomic contain or do peaks. The main idea to manually label small subset the genome, then learn...
In a world with data that change rapidly and abruptly, it is important to detect those changes accurately. this paper we describe an R package implementing generalized version of algorithm recently proposed by Hocking, Rigaill, Fearnhead, Bourque (2020) for penalized maximum likelihood inference constrained multiple change-point models. This can be used pinpoint the precise locations abrupt in large sequences. There are many application domains such models, as medicine, neuroscience or...
Neuroblastoma, a pediatric tumor of the sympathetic nervous system, is predominantly driven by copy number aberrations, which predict survival outcome in global neuroblastoma cohorts and low-risk cases. For high-risk patients there still need for better prognostic biomarkers. Via an international collaboration, we collected profiles 556 neuroblastomas generated on different array platforms. This manuscript describes composition dataset, methods used to process data, including segmentation...