Site Glossary

Inference Types:

Two inference types are available. The standard, and more reliable, inference considers only taxa for which chromosome number information is available. The extended inference includes taxa for which chromosome number information is not available, and offers ploidy inference for such taxa based on their siblings for which chromosome number data is available. Mind that ploidy inference for taxa with no chromosome number information is only possible for polyploidy inference, under the assumption that disploidization is not likely to occur in the time-scales of the examined phylogenies. As such, ploidy inference for taxa with no chromosome number information is biased towards polyploidy.

Ploidy Estimates Criteria:

In this section we cover to criteria that were used in the ploidy estimates tables:

• The Ploidy classification reliability score –
  This value corresponds to the normalized distance of the inferred polyploidy frequency from the threshold used to determine ploidy level. Specifically, ploidy classification of a taxon is done based on the frequency of its inference as polyploid across the sampled stochastic mappings, that represent different evolutionary histories of chromosome number that lead to the observed chromosome numbers at the tip taxa. Here, we used to two thresholds for classification of a taxon as either polyploid of diploid. If the frequency at which the taxon was inferred as polyploid across mappings is below 0.25, we classify it as a diploid. If the frequency at which it was classified as polyploid is above 0.75, we classify it as a polyploid. If the frequency falls between the two thresholds – we declare the ploidy level as uncertain. If the ploidy level is determined, the reliability of the classification can be measured as the normalized distance of the estimated polyploidy frequency from the used threshold used. For example, if the polyploidy frequency of a taxon is 0.85, then its distance from the used threshold, 0.75, is 0.1. The normalized distance is accordingly 0.4. in such case, we say that the classification reliability score is 0.4. However, if the polyploidy frequency is 0 (i.e., the taxon was classified as diploid across all the sampled mappings), then its distance from the used threshold, 0.25, is 0.25. Accordingly, the classification reliability score is 1.


• The Phylogenetically reliable flag –
  This binary value indicates whether ploidy classification of the taxon was different from the reported ploidy level when using an alternative phylogeny for ploidy inference. For each clade at a taxonomic level (i.e., genus or family), we reconstructed species trees using the One Two Tree software. The then applied our ploidy inference scheme on the reconstructed trees and extracted the obtained ploidy inferences. If, for a given taxon, the ploidy inference that is based on the original tree, partitioned from the ALLMB broad seed plant phylogeny, is different from the ploidy inference based on the reconstructed tree, we flag the ploidy inference for that taxon as unreliable.