https://github.com/bojarlab/gifflar

Glycan Informed Foundational Framework for Learning Abstract Representations, based on Combinatorial Complexes and Heterogeneous GNNs

https://github.com/bojarlab/gifflar

Science Score: 36.0%

This score indicates how likely this project is to be science-related based on various indicators:

  • CITATION.cff file
  • codemeta.json file
    Found codemeta.json file
  • .zenodo.json file
    Found .zenodo.json file
  • DOI references
  • Academic publication links
    Links to: arxiv.org
  • Academic email domains
  • Institutional organization owner
  • JOSS paper metadata
  • Scientific vocabulary similarity
    Low similarity (8.5%) to scientific vocabulary

Keywords

combinatorial-complex foundational-models glycan glycobiology graph-neural-network heterogeneous-graph-neural-network pretraining
Last synced: 6 months ago · JSON representation

Repository

Glycan Informed Foundational Framework for Learning Abstract Representations, based on Combinatorial Complexes and Heterogeneous GNNs

Basic Info
  • Host: GitHub
  • Owner: BojarLab
  • Language: HTML
  • Default Branch: main
  • Homepage:
  • Size: 2.05 MB
Statistics
  • Stars: 3
  • Watchers: 1
  • Forks: 2
  • Open Issues: 0
  • Releases: 0
Topics
combinatorial-complex foundational-models glycan glycobiology graph-neural-network heterogeneous-graph-neural-network pretraining
Created over 1 year ago · Last pushed 9 months ago
Metadata Files
Readme

README.md

GIFFLAR - Glycan-Informed Foundational Framework to Learn Abstract Representations

Glycans are the most complex biological sequence, with monosaccharides forming extended, non-linear sequences. As post-translational modifications, they modulate protein structure, function, and interactions. Due to their diversity and complexity, predictive models of glycan properties and functions are still insufficient. Graph Neural Networks (GNNs) are deep learning models designed to process and analyze graph-structured data. These architectures leverage the connectivity and relational information in graphs to learn effective representations of nodes, edges, and entire graphs. Iteratively aggregating information from neighboring nodes, GNNs capture complex patterns within graph data, making them particularly well-suited for tasks such as link prediction or graph classification across domains. This work presents a new model architecture based on combinatorial complexes and higher-order message passing to extract features from glycan structures into a latent space representation. The architecture is evaluated on an improved GlycanML benchmark suite, establishing a new state-of-the-art performance. We envision that these improvements will spur further advances in computational glycosciences and reveal the roles of glycans in biology.

arXiv

Installation:

bash conda create -n gifflar -y python=3.11 conda activate gifflar pip install -r requirements.txt

Owner

  • Name: BojarLab
  • Login: BojarLab
  • Kind: organization
  • Email: daniel.bojar@gu.se
  • Location: Gothenburg, Sweden

Machine Learning in Glycobiology and Systems Biology

GitHub Events

Total
  • Watch event: 2
  • Delete event: 2
  • Push event: 33
  • Fork event: 3
Last Year
  • Watch event: 2
  • Delete event: 2
  • Push event: 33
  • Fork event: 3

Dependencies

.github/workflows/test.yaml actions
  • actions/checkout v4 composite
  • conda-incubator/setup-miniconda v3 composite
requirements.txt pypi
  • jsonargparse *
  • networkx *
  • numpy *
  • pandas *
  • pytest *
  • pytorch-lightning *
  • pyyaml *
  • rdkit >=2022
  • rich *
  • scikit-learn *
  • torch *
  • torch_geometric *
  • torchmetrics *