Microbiome Maps: Hilbert Curve Visualizations of Metagenomic Profiles

Valdes, Camilo; Stebliankin,; Ruiz-Perez, Daniel; Ji, Park; H, Lee; Narasimhan, Giri

doi:10.1101/2021.03.22.436520

Cited by 1 publication

(1 citation statement)

References 46 publications

(51 reference statements)

Supporting

Mentioning

Contrasting

Order By: Relevance

“…1), where each channel is dedicated to a feature type. Transforming data into 2D images (i.e., feature maps) prior to applying the ML model was successfully used for other ML applications by Valdes et al [65]. Such transformations work well because ML techniques like convolutions work best on images.…”

Section: Methodsmentioning

confidence: 99%

PIsToN: Evaluating Protein Binding Interfaces with Transformer Networks

Stebliankin

Shirali

Baral

et al. 2023

Preprint

Self Cite

View full text Add to dashboard Cite

The computational studies of protein binding are widely used to investigate fundamental biological processes and facilitate the development of modern drugs, vaccines, and therapeutics. Scoring functions aim to predict complexes that would be formed by the binding of two biomolecules and to assess and rank the strength of the binding at the interface. Despite past efforts, the accurate prediction and scoring of protein binding interfaces remain a challenge. The physics-based methods are computationally intensive and often have to trade accuracy for computational cost. The possible limitations of current machine learning (ML) methods are ineffective data representation, network architectures, and limited training data. Here, we propose a novel approach called PIsToN (evaluating Protein binding Interfaces with Transformer Networks) that aim to distinguish native-like protein complexes from decoys. Each protein interface is transformed into a collection of 2D images (interface maps), where each im- age corresponds to a geometric or biochemical property in which pixel intensity represents the feature values. Such a data representation provides atomic-level resolution of relevant protein characteristics. To build hybrid machine learning models, additional empirical-based energy terms are computed and provided as inputs to the neural network. The model is trained on thousands of native and computationally-predicted protein complexes that contain challenging examples. The multi-attention transformer network is also endowed with explainability by highlighting the specific features and binding sites that were the most important for the classification decision. The developed PIsToN model significantly outperforms existing state-of-the-art scoring functions on well-known datasets.

show abstract

Section: Methodsmentioning

confidence: 99%