BIOLOGICAL FUNCTIONAL ANALYSIS IN THE AGE OF AI - WEBINAR

To whom: Academics & researchers, undergraduate & postgraduate students

When (time and date): 22.10.2025 at 16.00 EEST

Where: Online (MS Teams)

How to participate (registration link and DL): https://forms.gle/cVni5QzncHAmVmuX7

(A link to access the event will be sent to all registered participants upon successful registration.)

Description of the event:

Talk about the main rules to apply AI for biological functional Analysis

Biological functional analysis is important in environmental biology. The vast sequencing of microbial communities from soil, oceans, and extreme environments produces data at a scale where traditional annotation methods are insufficient.

AI now helps us predict microbial functions from metagenomic datasets, identifying genes responsible for carbon cycling, methane production, or nitrogen fixation. Another example is environmental DNA (eDNA) analysis: AI models can classify and functionally interpret DNA fragments from rivers or oceans, allowing us to monitor biodiversity loss, invasive species, or ecosystem resilience.

In this way, AI-powered functional analysis is helping us link molecular biology with global environmental challenges — from climate change to pollution control — turning raw sequencing data into actionable knowledge for sustainability

Learning goals and competencies:

The participants will learn

What is Biological Functional Analysis?

• Definition: the process of uncovering the biological meaning behind lists of genes, proteins, or metabolites.
• Classic approaches: Gene Ontology (GO), KEGG pathways, Reactome.
• Limitations: static databases, empirical methods, difficulty handling complex networks or context-specific data.

The Role of AI and how to apply?

• Data Quality: Environmental data (e.g., soil, ocean, air microbiomes) are often noisy, incomplete, or unevenly sampled.
• Integration of Heterogneous Data : Environmental studies combine genomics, climate data, chemical measurements, satellite imaging, etc
• Scale and Context : Functions learned in one environment (e.g., Arctic microbes) may not generalize to another (e.g., tropical soils).
• Data quality and bias: issues arising from incomplete or skewed datasets.
• Reproducibility: the need for open data, transparent methods, and reproducible pipelines.

Organizing HEI: UTH

Responsible person: Prof. Artemis Hatzigeorgiou (arhatzig@uth.gr)

Download calendar file (.ics)