Brewery

Brewery predicts protein structural annotations (PSAs) from amino acid sequences using ab initio, evolutionary-informed methods.


Key Features:

  • Ab Initio Prediction: Operates independently of template-based methods to predict PSAs for proteins lacking homologous structures.
  • Use of Evolutionary Data: Integrates multiple sources of evolutionary information and intrinsic sequence-derived properties to inform predictions.
  • Secondary Structure Prediction: Predicts secondary structure elements such as alpha-helices and beta-sheets.
  • Structural Motifs Identification: Identifies recurring structural motifs relevant to protein function and interactions.
  • Relative Solvent Accessibility: Estimates residue solvent exposure to inform stability and interaction analyses.
  • Contact Density Prediction: Predicts contact density maps to provide insights into residue spatial arrangements.

Scientific Applications:

  • Three-dimensional structure prediction: Supports inferring 3D structures from amino acid sequences for proteins without structural templates.
  • Novel fold characterization: Enables study of proteins with novel folds absent from structural databases.
  • Functional and interaction analysis: Facilitates investigation of protein function and interaction networks through structural annotations.
  • Evolutionary studies: Assists analysis of evolutionary relationships via integration of evolutionary information in structural predictions.

Methodology:

Brewery applies template-independent ab initio prediction methods that rely on intrinsic sequence-derived properties and integration of multiple evolutionary datasets to generate PSAs including secondary structure, structural motifs, relative solvent accessibility, and contact density maps.

Topics

Details

License:
CC-BY-NC-SA-4.0
Tool Type:
command-line tool, library
Programming Languages:
Python
Added:
1/18/2021
Last Updated:
2/6/2021

Operations

Publications

Torrisi M, Pollastri G. Brewery: deep learning and deeper profiles for the prediction of 1D protein structure annotations. Bioinformatics. 2020;36(12):3897-3898. doi:10.1093/bioinformatics/btaa204. PMID:32207516.

PMID: 32207516
Funding: - Irish Research Council: GOIPG/2015/3717

Downloads