Journal: mSystems

Article Title: Pangenome-scale annotation of mycobacteriophages for dissecting phage–host interactions based on a sequence clustering and structural homology analysis strategy

doi: 10.1128/msystems.00508-25

Figure Lengend Snippet: SCSH workflow and clusters summary. ( a ) A three-step SCSH workflow was used to annotate mycobacteriophages. First, MMseqs2 clustered 240,754 mycobacteriophage proteins with 50% sequence similarity and 90% sequence overlap, reducing the database to 14,622 clusters. Subsequently, AlphaFold predicted the structure of the representative sequence for each cluster. Finally, Foldseek was utilized to identify the optimal match for each representative structure in PDB, AFDB_SwissProt, and AFDB. ( b ) Summary of cluster distributions for different cluster sizes. ( c ) mycoPHG_clu_DB structural and Pfam consistency. The clusters exhibit a median LDDT of 0.923 and a median template modeling (TM) score of 0.907. Among all clusters, 94.8% with Pfam annotations exhibit 100% consistency. ( d ) Summary of sequence-based and Foldseek-based annotation rates across different phage clusters is presented in three panels. The left panel illustrates the average sequence annotation rate (34.00%) and the average structural annotation rate (52.11%) for 2,169 mycobacteriophages. The upper right panel displays the average structural-based annotation rate for each phage cluster, while the lower right panel shows the average sequences-based annotation rate for each phage cluster.

Article Snippet: All data used for this study are publicly available in AFDB ( https://alphafold.ebi.ac.uk/ , v.4, with specific examples corresponding to UniProt IDs Q9P1W9 , A0A6N2GQB0, A0AX0GGC2, and A0A2M9FCT4), the CATH database ( https://www.cathdb.info/ , v.4.3.0), the RCSB PDB ( https://www.rcsb.org/ , PDB IDs 4NUS, 2I91, 6MDX, and 4O16), the Foldseek database ( https://foldseek.steineggerlab.workers.dev/ , pre-generated Alphafold/UniProt, Alphafold/Swiss-Prot, and PDB100 databases), and National Center for Biotechnology Information Assembly Database ( https://www.ncbi.nlm.nih.gov/assembly/ , see for the EntrezIDs).

Techniques: Sequencing

Journal: mSystems

Article Title: Pangenome-scale annotation of mycobacteriophages for dissecting phage–host interactions based on a sequence clustering and structural homology analysis strategy

doi: 10.1128/msystems.00508-25

Figure Lengend Snippet: Analysis of structure without homologs uncovers novel enzymes and folds. ( a ) Summary of average reference structure pLDDT distributions for different dark cluster sizes (left): from left to right, each bin’s median pLDDT is 70.61, 69.54, 70.96, 73.01, 70.02, 72.24, 74.63, 76.18, 72.46. Summary of the percentages of High_pLDDT and Low_pLDDT proteins in each cluster (right). Proteins with an average pLDDT score ≥70 were defined as High_pLDDT, while those with a score <70 were defined as Low_pLDDT. ( b ) The AlphaFold model depicts the near-full helix structure encoded by the mycobacteriophage, which lacks a global match in PDB, AFDB_SwissProt, and AFDB. ( c ) The AlphaFold model of the three-domain protein encoded by the mycobacteriophage does not yield global matches in PDB, AFDB_SwissProt, and AFDB. Nevertheless, its three structural domains exhibit near-exact matches. ( d ) The AlphaFold model of the helix-turn-helix structure encoded by the mycobacterium phage aligns with entries in databases associated with diverse biological processes, such as anti-CRISPR, anti-RecBCD, and anti-population sensing. ( e ) Counts of EC number most frequently predicted by DeepFRI on the set of 826 cluster representative structures with an average pLDDT score above 90. ( f ) Two examples of novel enzymes predicted by DeepFRI (MPC00070, MPC01523) exhibit predicted pockets highlighted in light yellow.

Article Snippet: All data used for this study are publicly available in AFDB ( https://alphafold.ebi.ac.uk/ , v.4, with specific examples corresponding to UniProt IDs Q9P1W9 , A0A6N2GQB0, A0AX0GGC2, and A0A2M9FCT4), the CATH database ( https://www.cathdb.info/ , v.4.3.0), the RCSB PDB ( https://www.rcsb.org/ , PDB IDs 4NUS, 2I91, 6MDX, and 4O16), the Foldseek database ( https://foldseek.steineggerlab.workers.dev/ , pre-generated Alphafold/UniProt, Alphafold/Swiss-Prot, and PDB100 databases), and National Center for Biotechnology Information Assembly Database ( https://www.ncbi.nlm.nih.gov/assembly/ , see for the EntrezIDs).

Techniques: CRISPR

Journal: Current Issues in Molecular Biology

Article Title: Identification of SmNAC28 Transcription Factor and Its Mechanism of Regulating Salt Tolerance in Eggplant via S-Palmitoylation

doi: 10.3390/cimb48040398

Figure Lengend Snippet: Subcellular localization and structure of SmNAC28 . ( A ) Subcellular localization of 35S- SmNAC28 -GFP and the double mutant 35S- SmNAC28 C25+28S -GFP. The 35S-GFP construct served as the control. (Scale bars = 20 μm). Using plasma membrane (PM) and nucleus (NC) markers as reference ( B ) AlphaFold-predicted protein structure of SmNAC28 . The overall protein structure is shown in red; the inset highlights a zoomed-in view of the region containing residues C25 and C28. ( C ) Effect of stress treatment on SmNAC28 localization. GFP fluorescence, bright-field, and merged images of 35S- SmNAC28 -GFP under Mock, NH 2 OH (hydroxylamine treatment), and Salt (100 mM NaCl) conditions are presented. Arrows indicate the relocalization of SmNAC28 from the membrane to the nucleus following NH 2 OH treatment. (Scale bars = 20 μm).

Article Snippet: The tertiary structures of eggplant NAC proteins were predicted using the AlphaFold Protein Structure Database (DeepMind, Islington, London, UK) ( https://alphafold.com/ , accessed on 25 December 2025).

Techniques: Mutagenesis, Construct, Control, Clinical Proteomics, Membrane, Fluorescence