Extended Data Fig. 10: Domains in N-terminal regions of SNIPE homologs.
From: A membrane-bound nuclease directly cleaves phage DNA during genome injection
(a) Structures of the 34% of SNIPE homologs that lack a predicted TM were predicted by AlphaFold2 and the percentages of homologs with a globular domain at their N-terminus (PH or DUF4428) are shown. Homology was detected by FoldSeek and/or HHpred. (b) Example of a predicted SNIPE homolog structure generated by AlphaFold2. The structure lacks both a predicted transmembrane (TM) domain and a globular domain in the N-terminal region. (c) Example of a predicted SNIPE homolog structure generated by AlphaFold2. The structure lacks a predicted transmembrane (TM) domain and features a predicted pleckstrin homology (PH) domain at the N-terminus. Top matches from FoldSeek (using the PDB100 database of experimentally determined structures) and an HHpred search with the N-terminal globular domain are presented. (d) Example of a predicted SNIPE homolog structure generated by AlphaFold2. The structure lacks a predicted transmembrane (TM) domain and features a predicted DUF4428 domain at the N-terminus. The top match from an HHpred search with the N-terminal globular domain is presented. (e) Structural alignment between the predicted PH domain of SNIPE homolog KEA32402.1 and the PH domain of GSK2 (PDB 2BCJ-A). (f) Same insets as shown in Fig. 5f, but with the percentage of different residues found at a given location across SNIPE homologs, as quantified by ConSurf. Only the top three most abundant residues are shown for each location. Positively charged amino acids are labeled in blue for the top inset. Mutations that enhanced defense against Bas14 are marked in red for the bottom inset.
