Multi-layer Trajectory Clustering: a Network Algorithm for Disease Subtyping

Structural and functional multilayer network analysis in restless legs syndrome patients

The combination of brain structural and functional connectivity offers complementary insights into its organisation. Multilayer network analysis explores various relationships across different layers within a single system. We aimed to investigate changes in the structural and functional multilayer network in 69 patients with primary restless legs syndrome (RLS) compared with 50 healthy controls. Participants underwent diffusion tensor imaging (DTI) and resting state-functional magnetic resonance imaging (rs-fMRI) using a three-tesla MRI scanner. We constructed a structural connectivity matrix derived from DTI using a DSI program and made a functional connectivity matrix based on rs-fMRI using an SPM program and CONN toolbox. A multilayer network analysis, using BRAPH program, was then conducted to assess the connectivity patterns in both groups. At the global level, significant differences there were between the patients with RLS and healthy controls. The average multiplex participation was lower in patients with RLS than in healthy controls (0.804 vs. 0.821, p = 0.042). Additionally, several regions showed significant differences in the nodal level in multiplex participation between patients with RLS and healthy controls, particularly the frontal and temporal lobes. The regions affected included the inferior frontal gyrus, medial orbital gyrus, precentral gyrus, rectus gyrus, insula, superior and inferior temporal gyrus, medial and lateral occipitotemporal gyrus, and temporal pole. These results represent evidence of diversity in interactions between structural and functional connectivity in patients with RLS, providing a more comprehensive understanding of the brain network in RLS. This may contribute to a precise diagnosis of RLS, and aid the development of a biomarker to track treatment effectiveness.

The collective vs individual nature of mountaineering: a network and simplicial approach

Mountaineering is a sport of contrary forces: teamwork plays a large role in mental fortitude and skills, but the actual act of climbing, and indeed survival, is largely individualistic. This work studies the effects of the structure and topology of relationships within climbers on the level of cooperation and success. It does so using simplicial complexes, where relationships between climbers are captured through simplices that correspond to joint previous expeditions with dimension given by the number of climbers minus one and weight given by the number of occurrences of the simplex. First, this analysis establishes the importance of relationships in mountaineering and shows that chances of failure to summit reduce drastically when climbing with repeated partners. From a climber-centric perspective, it finds that climbers that belong to simplices with large dimension were more likely to be successful, across all experience levels. Then, the distribution of relationships within a group is explored to categorize collective human behavior in expeditions, on a spectrum from polarized to cooperative. Expeditions containing simplices with large dimension, and usually low weight (weak relationships), implying that a large number of people participated in a small number of joint expeditions, tended to be more cooperative, improving chances of success of all members of the group, not just those that were part of the simplex. On the other hand, the existence of small, usually high weight (i.e., strong relationships) simplices, subgroups lead to a polarized style where climbers that were not a part of the subgroup were less likely to succeed. Lastly, this work examines the effects of individual features (such as age, gender, climber experience etc.) and expedition-wide factors (number of camps, total number of days etc.) that are more important determiners of success in individualistic and cooperative expeditions respectively. Centrality indicates that individual features of youth and oxygen use while ascending are the most important predictors of success. Of expedition-wide factors, the expedition size and number of expedition days are found to be strongly correlated with success rate.

Stroke recovery phenotyping through network trajectory approaches and graph neural networks

Stroke is a leading cause of neurological injury characterized by impairments in multiple neurological domains including cognition, language, sensory and motor functions. Clinical recovery in these domains is tracked using a wide range of measures that may be continuous, ordinal, interval or categorical in nature, which can present challenges for multivariate regression approaches. This has hindered stroke researchers’ ability to achieve an integrated picture of the complex time-evolving interactions among symptoms. Here, we use tools from network science and machine learning that are particularly well-suited to extracting underlying patterns in such data, and may assist in prediction of recovery patterns. To demonstrate the utility of this approach, we analyzed data from the NINDS tPA trial using the Trajectory Profile Clustering (TPC) method to identify distinct stroke recovery patterns for 11 different neurological domains at 5 discrete time points. Our analysis identified 3 distinct stroke trajectory profiles that align with clinically relevant stroke syndromes, characterized both by distinct clusters of symptoms, as well as differing degrees of symptom severity. We then validated our approach using graph neural networks to determine how well our model performed predictively for stratifying patients into these trajectory profiles at early vs. later time points post-stroke. We demonstrate that trajectory profile clustering is an effective method for identifying clinically relevant recovery subtypes in multidimensional longitudinal datasets, and for early prediction of symptom progression subtypes in individual patients. This paper is the first work introducing network trajectory approaches for stroke recovery phenotyping, and is aimed at enhancing the translation of such novel computational approaches for practical clinical application.