Figure 1.
Critical points of the L-curve.
The critical point represents a 90% reduction in the maximum variation value, and points above the critical point correspond to key K-strings. The curves were generated based on randomly selected datasets with different numbers of species, (20, 40, 80, 100, or 200). Points with serial numbers greater than 35,000 do not appear under zoom.
Figure 2.
Phylogenetic trees constructed from our key K-strings and complete K-strings using the CVTree method.
Both trees contain 87 species that were randomly selected from each phylum (Table S2). (a) The phylogenetic trees constructed from our 23,223 key K-strings; (b) The phylogenetic trees constructed from the complete set of 205 K-strings using the CVTree method. Bootstrap support values above 40% from 100 replicates were show in the figure.
Table 1.
The number of phylum-specific key K-strings and their diversity.a
Figure 3.
Single amino acid compositions of the key K-strings in each phylum.
The compositional patterns of 8 groups of phylum-specific key K-strings are illustrated using different colors.
Figure 4.
The composition patterns of phylum-specific key K-strings in different groups.
The 8 phyla are arranged according to the traditional view of phylogeny: Porifera, Coelenterata, Platyhelminthes, Nematoda, Annelida, Crustacea, Echinodermata and Vertebrata. Single amino acids, dimers and triplets containing the amino acids A, P or T displayed one distribution pattern, whereas those containing C exhibited a different distribution pattern.
Figure 5.
Distribution of conserved regions and regions in which key K-strings accumulated for complete protein sequences.
Thirteen protein sequences are aligned in order and linked head-to-tail for the 10 chosen vertebrates. The curve models the distribution of sequence conservativity, and gray regions represent regions in which key K-strings accumulate.