# Skin microbiome

Reference genomes for the human skin microbiome have been scarce, especially outside bacteria and across global strain variation. The Global Skin Metagenomic Data resource combined 12,409 newly sequenced and 4,396 public skin metagenomes with 3,077 isolate genomes to reconstruct 452,397 metagenome-assembled genomes and add 42,758 previously uncharacterized taxa. From 1.71 billion predicted protein sequences it consolidated 1,272,886 protein families, 79,434 of which had no homology hit in Pfam or RefSeq.

Reference genomes for the human skin microbiome have been scarce, especially outside bacteria and across global strain variation. The Global Skin Metagenomic Data resource combined 12,409 newly sequenced and 4,396 public skin metagenomes with 3,077 isolate genomes to reconstruct 452,397 metagenome-assembled genomes and add 42,758 previously uncharacterized taxa. From 1.71 billion predicted protein sequences it consolidated 1,272,886 protein families, 79,434 of which had no homology hit in Pfam or RefSeq.

This review summarizes the established clinical framework for atopic dermatitis (AD) diagnosis alongside emerging tools intended to complement it—circulating and skin-derived biomarkers, minimally invasive tape stripping, the skin microbiome, instrumental barrier assessment, noninvasive biofluids, advanced optical imaging (reflectance confocal microscopy, optical coherence tomography and line-field confocal OCT), and artificial-intelligence and digital tools, with attention to special populations—concluding that clinical diagnosis remains the reference standard, that the experienced clinician still outperforms any single test, and that these tools are promising adjuncts though few are standardized or prospectively validated.

This review summarizes the established clinical framework for atopic dermatitis (AD) diagnosis alongside emerging tools intended to complement it—circulating and skin-derived biomarkers, minimally invasive tape stripping, the skin microbiome, instrumental barrier assessment, noninvasive biofluids, advanced optical imaging (reflectance confocal microscopy, optical coherence tomography and line-field confocal OCT), and artificial-intelligence and digital tools, with attention to special populations—concluding that clinical diagnosis remains the reference standard, that the experienced clinician still outperforms any single test, and that these tools are promising adjuncts though few are standardized or prospectively validated.