Correspondence analysis as a tool in fungal taxonomy

Area of Science:

• Taxonomy
• Numerical Taxonomy
• Bioinformatics

Background:

• Polyphasic taxonomy generates complex datasets requiring numerical analysis.
• Mixed-type data presents mathematical challenges in taxonomic studies.
• Correspondence analysis (CA) is underutilized in taxonomy despite its utility.

Purpose of the Study:

• To explore the application of Correspondence Analysis (CA) in solving taxonomic problems.
• To demonstrate CA's capability in handling mixed-type data and large datasets.
• To highlight CA's potential for identifying taxonomic relationships and aiding in identification.

Main Methods:

• Application of Correspondence Analysis (CA), an ordination technique.
• Coding of mixed-type data for CA compatibility.
• Simultaneous ordination of variables and taxa.

Main Results:

• CA effectively analyzes taxonomic interrelationships between taxa and variables.
• The method is suitable for large and small datasets, including those with missing data.
• CA aids in screening datasets, identifying discriminating variables, detecting outliers, and facilitating sample identification.

Conclusions:

• Correspondence Analysis (CA) is a versatile tool for taxonomic research, particularly with mixed-type data.
• CA assists in constructing identification keys and grouping variables by discriminatory importance.
• The technique's sensitivity to outliers can be leveraged for data verification.