Dataset Information

PRER: A patient representation with pairwise relative expression of proteins on biological networks.

ABSTRACT: Changes in protein and gene expression levels are often used as features in predictive modeling such as survival prediction. A common strategy to aggregate information contained in individual proteins is to integrate the expression levels with the biological networks. In this work, we propose a novel patient representation where we integrate proteins' expression levels with the protein-protein interaction (PPI) networks: Patient representation with PRER (Pairwise Relative Expressions with Random walks). PRER captures the dysregulation patterns of proteins based on the neighborhood of a protein in the PPI network. Specifically, PRER computes a feature vector for a patient by comparing the source protein's expression level with other proteins' levels that are within its neighborhood. The neighborhood of the source protein is derived by biased random-walk strategy on the network. We test PRER's performance in survival prediction task in 10 different cancers using random forest survival models. PRER yields a statistically significant predictive performance in 9 out of 10 cancers when compared to the same model trained with features based on individual protein expressions. Furthermore, we identified the pairs of proteins that their interactions are predictive of patient survival but their individual expression levels are not. The set of identified relations provides a valuable collection of protein biomarkers with high prognostic value. PRER can be used for other complex diseases and prediction tasks that use molecular expression profiles as input. PRER is freely available at: https://github.com/hikuru/PRER.

SUBMITTER: Kuru HI

PROVIDER: S-EPMC8238204 | biostudies-literature |

REPOSITORIES: biostudies-literature

ACCESS DATA

Similar Datasets

Project description:BackgroundEcological communities of interacting species analyzed as complex networks have shown that species dependence on their counterparts is more complex than expected at random. As for other potentially mutualistic interactions, ant-plant networks mediated by extrafloral nectar show a nested (asymmetric) structure with a core of generalist species dominating the interaction pattern. Proposed factors structuring ecological networks include encounter probability (e.g., species abundances and habitat heterogeneity), behavior, phylogeny, and body size. While the importance of underlying factors that influence the structure of ant-plant networks have been separately explored, the simultaneous contribution of several biological and ecological attributes inherent to the species, guild or habitat level has not been addressed.MethodsFor a tropical seasonal site we recorded (in 48 censuses) the frequency of pairwise ant-plant interactions mediated by extrafloral nectaries (EFN) on different habitats and studied the resultant network structure. We addressed for the first time the role of mechanistic versus neutral determinants at the 'fine-grain' structure (pairwise interactions) of ant-plant networks. We explore the simultaneous contribution of several attributes of plant and ant species (i.e., EFN abundance and distribution, ant head length, behavioral dominance and invasive status), and habitat attributes (i.e., vegetation structure) in prevailing interactions as well as in overall network topology (community).ResultsOur studied network was highly-nested and non-modular, with core species having high species strengths (higher strength values for ants than plants) and low specialization. Plants had higher dependences on ants than vice versa. We found that habitat heterogeneity in vegetation structure (open vs. shaded habitats) was the main factor explaining network and fine-grain structure, with no evidence of neutral (abundance) effects.DiscussionCore ant species are relevant to most plants species at the network showing adaptations to nectar consumption and deterrent behavior. Thus larger ants interact with more plant species which, together with higher dependence of plants on ants, suggests potential biotic defense at a community scale. In our study site, heterogeneity in the ant-plant interactions among habitats is so prevalent that it emerges at community-level structural properties. High frequency of morphologically diverse and temporarily-active EFNs in all habitats suggests the relevance and seasonality of plant biotic defense provided by ants. The robust survey of ecological interactions and their biological/ecological correlates that we addressed provides insight of the interplay between adaptive-value traits and neutral effects in ecological networks.

Dataset Information

PRER: A patient representation with pairwise relative expression of proteins on biological networks.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets