Glycowork: A Python package for glycan data science and machine learning.

Area of Science:

• Biochemistry
• Computational Biology
• Glycoscience

Background:

• Glycans are vital for numerous biological functions but are challenging to analyze computationally.
• Current analytical methods require significant computational expertise, limiting their use by many researchers.

Purpose of the Study:

• To introduce glycowork, an open-source Python package designed to simplify glycan data science and machine learning.
• To empower researchers with limited computational backgrounds to integrate glycan analysis into their workflows.

Main Methods:

• Development of an open-source Python package, glycowork.
• Implementation of functions for automatic glycan motif annotation and distribution analysis (heatmaps, statistical enrichment).
• Inclusion of visualization tools, database interaction routines, and pre-trained machine learning models.

Main Results:

• Glycowork provides accessible tools for analyzing glycan motifs and their distributions.
• The package facilitates the use of machine learning for glycan data science.
• Demonstrated utility through workflows analyzing glycan motifs in diverse biological contexts.

Conclusions:

• Glycowork lowers the barrier to entry for computational glycan analysis.
• The package enables researchers to extract novel insights from glycan datasets.
• Glycowork is freely available, promoting wider adoption in biological research.