Related Experiment Videos
A model of immunological distances in systematics
Journal of Theoretical Biology
|June 21, 1985
Summary
Immunological distances in systematics can be unreliable due to non-metricity. This study presents a model and methods to transform these distances into metric ones for accurate phylogenetic analysis.
Area of Science:
- Systematics
- Immunology
- Phylogenetics
Background:
- Immunological distances are widely used in systematics.
- Concerns exist regarding their non-metric nature, specifically non-reciprocity and violation of triangle inequality.
- This non-metricity questions their utility in phylogenetic analysis.
Purpose of the Study:
- To present a model relating observed immunological distance to antigenic site differences.
- To develop procedures for transforming non-metric immunological distances into metric distances.
- To address scaling and non-reciprocity issues in distance matrices for improved phylogenetic inference.
Main Methods:
- A mathematical model was developed to link immunological distance with antigenic site differences.
- Two transformation procedures were proposed to convert immunological distances to metric distances.
- An extension of the ADCLUS procedure was suggested for further data transformation.
Main Results:
- The model explains departures from metric conditions (reciprocity, triangle inequality).
- Standard scaling adjustments for immunological distance matrices were found inappropriate.
- A proposed transformation is effective for similarity matrices, but non-reciprocity remains a challenge.
Conclusions:
- The developed model and transformations offer a strategy to derive metric distances from non-metric immunological data.
- This approach enhances the utility of immunological data for robust phylogenetic reconstruction.
- Further methodological extensions are suggested to fully address non-metricity issues.