Samy Coulombe’s research while affiliated with McGill University and other places

Dynamic graphs are rich data structures that are used to model complex relationships between entities over time. In particular, anomaly detection in temporal graphs is crucial for many real world applications such as intrusion identification in network systems, detection of ecosystem disturbances and detection of epidemic outbreaks. In this paper, we focus on change point detection in dynamic graphs and address three main challenges associated with this problem: i) how to compare graph snapshots across time, ii) how to capture temporal dependencies, and iii) how to combine different views of a temporal graph. To solve the above challenges, we first propose Laplacian Anomaly Detection (LAD) which uses the spectrum of graph Laplacian as the low dimensional embedding of the graph structure at each snapshot. LAD explicitly models short term and long term dependencies by applying two sliding windows. Next, we propose MultiLAD, a simple and effective generalization of LAD to multi-view graphs. MultiLAD provides the first change point detection method for multi-view dynamic graphs. It aggregates the singular values of the normalized graph Laplacian from different views through the scalar power mean operation. Through extensive synthetic experiments, we show that i) LAD and MultiLAD are accurate and outperforms state-of-the-art baselines and their multi-view extensions by a large margin, ii) MultiLAD’s advantage over contenders significantly increases when additional views are available, and iii) MultiLAD is highly robust to noise from individual views. In five real world dynamic graphs, we demonstrate that LAD and MultiLAD identify significant events as top anomalies such as the implementation of government COVID-19 interventions which impacted the population mobility in multi-view traffic networks.

HOTAIRM1 is unlike most long non-coding RNAs in that its sequence is highly conserved across mammals. Such evolutionary conservation points to it having a role in key cellular processes. We previously reported that HOTAIRM1 is required to curb premature activation of downstream HOXA genes in a cell model recapitulating their sequential induction during development. We found that it regulates 3’ HOXA gene expression by a mechanism involving epigenetic and three-dimensional chromatin changes. Here we show that HOTAIRM1 participates in proper progression through the early stages of neuronal differentiation. We found that it can associate with the HOXA1 transcription factor and contributes to its downstream transcriptional program. Particularly, HOTAIRM1 affects the NANOG/POU5F1/SOX2 core pluripotency network maintaining an undifferentiated cell state. HOXA1 depletion similarly perturbed expression of these pluripotent factors, suggesting that HOTAIRM1 is a modulator of this transcription factor pathway. Also, given that binding of HOTAIRM1 to HOXA1 was observed in different cell types and species, our results point to this ribonucleoprotein complex as an integral part of a conserved HOTAIRM1-HOXA1 regulatory axis modulating the transition from a pluripotent to a differentiated neuronal state.

Dynamic graphs are rich data structures that are used to model complex relationships between entities over time. In particular, anomaly detection in temporal graphs is crucial for many real world applications such as intrusion identification in network systems, detection of ecosystem disturbances and detection of epidemic outbreaks. In this paper, we focus on change point detection in dynamic graphs and address three main challenges associated with this problem: i). how to compare graph snapshots across time, ii). how to capture temporal dependencies, and iii). how to combine different views of a temporal graph. To solve the above challenges, we first propose Laplacian Anomaly Detection (LAD) which uses the spectrum of graph Laplacian as the low dimensional embedding of the graph structure at each snapshot. LAD explicitly models short term and long term dependencies by applying two sliding windows. Next, we propose MultiLAD, a simple and effective generalization of LAD to multi-view graphs. MultiLAD provides the first change point detection method for multi-view dynamic graphs. It aggregates the singular values of the normalized graph Laplacian from different views through the scalar power mean operation. Through extensive synthetic experiments, we show that i). LAD and MultiLAD are accurate and outperforms state-of-the-art baselines and their multi-view extensions by a large margin, ii). MultiLAD's advantage over contenders significantly increases when additional views are available, and iii). MultiLAD is highly robust to noise from individual views. In five real world dynamic graphs, we demonstrate that LAD and MultiLAD identify significant events as top anomalies such as the implementation of government COVID-19 interventions which impacted the population mobility in multi-view traffic networks.

HOTAIRM1 is unlike most long non-coding RNAs in that its sequence is highly conserved across mammals. Such evolutionary conservation points to it having a role in key cellular processes. We previously reported that HOTAIRM1 is required to curb premature activation of downstream HOXA genes in a cell model recapitulating their sequential induction during development. We found that it regulates 3' HOXA gene expression by a mechanism involving epigenetic and three-dimensional chromatin changes. Here we show that HOTAIRM1 is required for proper progression through the early stages of neuronal differentiation. We found that it associates with the HOXA1 transcription factor and participates in its downstream transcriptional program. Particularly, HOTAIRM1 affects the NANOG/POU5F1/SOX2 core pluripotency network maintaining an undifferentiated cell state. HOXA1 depletion similarly perturbed expression of these pluripotent factors, suggesting that HOTAIRM1 is a modulator of this transcription factor pathway. Also, given that binding of HOTAIRM1 to HOXA1 was observed in different cell types and species, our results point to this ribonucleoprotein complex as an integral part of a conserved HOTAIRM1-HOXA1 regulatory axis controlling the transition from a pluripotent to a differentiated neuronal state.