CLASV: Rapid Lassa virus lineage assignment with random forest.

Area of Science:

• Virology
• Computational Biology
• Epidemiology

Background:

• Lassa fever, caused by Lassa virus (LASV), is a severe hemorrhagic disease endemic to West Africa, causing significant mortality and morbidity.
• The virus affects approximately 300,000 individuals annually, resulting in about 5,000 deaths, and has spread globally through exported cases.
• Distinct LASV lineages exhibit varying immunological behaviors, necessitating rapid identification during outbreaks and for exported cases.

Purpose of the Study:

• To develop and present CLASV, a novel machine learning-based tool for the rapid and accurate assignment of Lassa virus lineages.
• To enable swift identification of circulating LASV lineages (II, III, and IV/V) from raw nucleotide sequences.

Main Methods:

• Development of CLASV utilizing a Random Forest classifier, a machine learning algorithm.
• Implementation of CLASV in Python for straightforward integration into existing bioinformatics workflows.
• Testing and validation of CLASV's performance in assigning LASV lineages.

Main Results:

• CLASV demonstrates rapid and accurate assignment of Lassa virus sequences to dominant circulating lineages (II, III, and IV/V).
• The tool processes raw nucleotide sequences efficiently, providing lineage identification crucial for epidemiological surveillance.
• CLASV is designed for ease of use and integration into public health and research settings.

Conclusions:

• CLASV provides a valuable computational tool for the rapid identification of Lassa virus lineages, supporting public health responses to Lassa fever outbreaks.
• Accurate lineage identification is critical for understanding disease epidemiology, transmission dynamics, and potentially informing targeted interventions.
• The freely available Python-based CLASV tool facilitates global efforts in Lassa fever surveillance and control.