We have applied a statistical protocol based on principal component analysis, clustering methods, and discriminant analysis for the identification of

Computer-assisted sperm analysis (CASA) allows assessing the motility of individual spermatozoa, generating huge datasets. These datasets can be analyzed using data mining techniques such as cluster analysis, to group the spermatozoa in subpopulations with biological meaning. This review considers the use of statistical techniques for clustering CASA data, their challenges and possibilities. There are many clustering approaches potentially useful for grouping sperm motility data, but some options may be more appropriate than others. Future development should focus not only in improvements of subpopulation analysis, but also in finding consistent biological meanings for these subpopulations.