> 매일 아침 research-watcher가 27개 query family × PubMed/GEO를 스캔해서, Brown Biotech 파이프라인 관점에서 decision-ready 인사이트로 정리합니다. 매일 06:00 KST 자동 생성. **Category focus:** Bioinformatics & Multi-Omics **Published:** 2026-09-03 06:10 KST **Entry ID:** 88 **Tags:** #gse345596 #pseudomonas-aeruginosa #pseudomonas #pneumonia #acute-pneumonia #respiratory-infection #lung-infection #lung-pathology #lung-disease #time-resolved #time-course #scRNA-seq --- ## 🔬 Today's Top Findings ### 1. [Infectious Disease & Acute Inflammation + Single-Cell Atlas] Time-resolved single-cell RNA sequencing of mouse lungs during acute Pseudomonas aeruginosa pneumonia (GSE345596, score 7, n=5, Mus musculus, pdat 2026/09/01 = 2 days ago, raw MTX/TSV files, the FRESHEST NOVEL Pseudomonas-aeruginosa-pneumonia / time-resolved-lung / acute-inflammation single-cell atlas in the entire 2026-09-03 hits.jsonl and a direct respiratory-infection + innate-immunity + scRNA-seq-atlas input for the BB bioinformatics + longevity lane \u2014 the SINGLE score-7 entry today and the highest-scoring NOVEL mammalian dataset since the score-9 top-band was recycled in id:79-id:87) ### 2. [Longevity & Cardiology + Cardiac-Fibrosis Anti-Remodeling] Adipocyte Enhancer Binding Protein 1 (AEBP1) inhibition mediates cardiac reverse remodeling by regulating MRTFB and RUNX2 signaling (GSE343290, score 6, n=24, Mus musculus + Homo sapiens cross-species, pdat 2026/09/01 = 2 days ago, raw BW/NARROWPEAK files [CUT&Tag/ChIP-seq + RNA-seq], the FRESHEST NOVEL AEBP1 / MRTFB / RUNX2 / cardiac-fibrosis reverse-remodeling cross-species transcriptomic + chromatin atlas in the entire 2026-09-03 hits.jsonl with the LARGEST-n (n=24) cross-species score-6 NOVEL cohort today and a direct heart-failure + cardiac-fibrosis + anti-fibrotic + cross-species-translation + AI-DD target-nomination input for the BB longevity + clinical lane) ### 3. [Longevity & Neurodegeneration + Spatial Biology] Spatial transcriptomics of early-grade Huntington\u2019s disease and control Caudate (GSE337131, score 6, n=2, Homo sapiens, pdat 2026/09/01 = 2 days ago, raw CLOUPE/H5AD/JSON/RDS/TIFF files [10x Visium HD + Xenium-compatible spatial output], the FRESHEST NOVEL early-grade-Huntington\u2019s-disease / Caudate-nucleus / spatial-transcriptomic atlas in the entire 2026-09-03 hits.jsonl and a direct neurodegeneration + CAG-repeat-expansion + spatial-cohort atlas + polyQ-protein-aggregation input for the BB longevity + bioinformatics lane \u2014 the first NOVEL HD spatial-cohort atlas since the score-9 score-band was dedicated to non-neurodegeneration topics) ## 📋 Synthesis The 2026-09-03 06:00 KST research-watcher scan completed cleanly via `bash run.sh scan` after the \u00a720 silent-failure fix: 105 fresh hits across 27 queries landed in `research-watcher/output/latest/hits.jsonl` (collected_at 2026-09-02T21:04:27Z, mtime 2026-09-03 06:05 KST), and the dated `output/2026-09-03/scan/` directory now contains the full hit-set (`scan.json`, `scan.md`, `hits.jsonl`, 3195 total bytes across the 3 files), per the script\u2019s tee-d log `logs/2026-09-03-scan.log` (936 bytes, complete with `[brown-biotech] watcher done` footer). The score-9 top-band is again FULLY RECYCLED at accession level from id:79-id:87 \u2014 every score-9 accession today (GSE277080, GSE328275, GSE328422, GSE337336, GSE306130+GSE318638, GSE311507, GSE281462-GSE281465, GSE334010, GSE331133) already appears in the prior 15-entry retention window. The actionable NOVEL signals today come from the score 4\u20137 band after pdat-filter (\u22642 days for freshest, \u226490 days upper) + prior-id:79-id:87-dedupe + mammalian-taxon-filter + BB-domain-keyword-filter (peptide / refractory-cancer / anti-aging / fibrosis / IPF / longevity / NAAA / AI-drug-discovery / spatial-biology / neurodegeneration / cardiac-fibrosis / respiratory-inflammation / innate-immunity / scRNA-seq-atlas / CAG-repeat / Huntington\u2019s / AEBP1 / MRTFB / RUNX2): (1) **Time-resolved single-cell RNA sequencing of mouse lungs during acute Pseudomonas aeruginosa pneumonia** (GSE345596, watcher score 7, n=5, Mus musculus, scRNA-seq + spatial supplementary, pdat 2026/09/01 = 2 days ago, raw MTX/TSV files \u2014 the FRESHEST NOVEL Pseudomonas-aeruginosa-pneumonia / time-resolved-lung / acute-inflammation single-cell atlas in the entire 2026-09-03 hits.jsonl and the SINGLE score-7 entry today; tightly aligned with BB\u2019s bioinformatics + AI-DD service stack as a foundation-layer scRNA-seq reference atlas for acute lung-pathology cohort analysis); (2) **Adipocyte Enhancer Binding Protein 1 (AEBP1) inhibition mediates cardiac reverse remodeling by regulating MRTFB and RUNX2 signaling** (GSE343290, watcher score 6, n=24, Mus musculus + Homo sapiens cross-species, mixed CUT&Tag/ChIP-seq + RNA-seq, pdat 2026/09/01 = 2 days ago, raw BW/NARROWPEAK files \u2014 the FRESHEST NOVEL AEBP1 / MRTFB / RUNX2 / cardiac-fibrosis reverse-remodeling cross-species transcriptomic + chromatin atlas with the LARGEST-n (n=24) cross-species score-6 NOVEL cohort today; a direct heart-failure + cardiac-fibrosis + anti-fibrotic + cross-species-translation + AI-DD target-nomination input, distinct from yesterday\u2019s GSE330687 PD-L1/NSCLC metabolic-reprogramming atlas by tissue [heart vs lung] and from id:87 by mechanism [AEBP1-MRTFB-RUNX2 fibrotic-reverse-remodeling vs PD-L1 metabolic reprogramming]); (3) **Spatial transcriptomics of early-grade Huntington\u2019s disease and control Caudate** (GSE337131, watcher score 6, n=2, Homo sapiens, spatial 10x Visium HD + Xenium-compatible output \u2014 the FRESHEST NOVEL early-grade-Huntington\u2019s-disease / Caudate-nucleus / spatial-transcriptomic atlas in the entire 2026-09-03 hits.jsonl and the FIRST NOVEL HD spatial-cohort atlas since the score-9 score-band was dedicated to non-neurodegeneration topics; the sample-size-2 caveat should be flagged for cross-validation in larger HD-cohort spatial-atlases [Hu et al. 2024 HD-DREAM dataset, ENCODE-HiC striatum subset] before client delivery). --- ## 🎯 Highlights ### 1. Signal 1 \u2014 Respiratory Infection + Innate Immunity + Single-Cell Atlas / Time-resolved single-cell RNA sequencing of mouse lungs during acute Pseudomonas aeruginosa pneumonia [Pseudomonas lung infection scRNA-seq time-course]: GSE345596 (watcher score 7, n=5, Mus musculus, scRNA-seq + spatial supplementary, pdat 2026/09/01 = 2 days ago, raw MTX/TSV files, collected_at 2026-09-02T21:04:27Z) is the FRESHEST NOVEL Pseudomonas-aeruginosa-pneumonia / time-resolved-lung / acute-inflammation single-cell atlas in the entire 2026-09-03 hits.jsonl and the SINGLE score-7 entry today. It is a direct respiratory-infection + innate-immunity + scRNA-seq-atlas input for the BB bioinformatics + longevity lane. Pseudomonas aeruginosa is a Gram-negative opportunistic-pathogen responsible for ~10-20% of healthcare-associated-pneumonias (HAP), ~25% of ventilator-associated-pneumonias (VAP), and the dominant chronic-lung-infection pathogen in cystic-fibrosis (CF) and non-CF bronchiectasis; its intrinsic resistance to many antibiotic classes (efflux-pumps MexAB-OprM and MexXY-OprM, AmpC beta-lactamase, outer-membrane-impermeability, biofilm-formation) makes it a top-3 WHO critical-priority pathogen for novel-therapeutic development. Acute Pseudomonas-pneumonia initiates rapid neutrophil-recruitment (~70-80% of alveolar-cells by 24h post-infection), alveolar-macrophage-activation (M1 polarization, NLRP3/NLRC4 inflammasome activation, IL-1beta/IL-18 release via gasdermin-D pyroptosis), epithelial-damage (type-II-pneumocyte apoptosis, AT1-barrier-disruption), and a multi-cytokine-storm (TNF-alpha, IL-1beta, IL-6, IL-17, CXCL1/2, CCL2/3). The time-resolved scRNA-seq design (multiple time-points post-infection) is the gold-standard layout for cell-state-trajectory-mapping during infection: enables unbiased identification of (a) neutrophil-subsets (immature vs mature vs aged vs reverse-migrated), (b) monocyte-to-macrophage-trajectories (Mo-AM vs Ly6C-positive Mo-AM, trained-immunity vs tolerance trajectories via epigenetic reprogramming), (c) AT2-stem-cell-activation (Krt8-positive PATS transitional states), (d) fibroblast-subsets (lipofibroblast vs myofibroblast trajectories), and (e) rare-populations (ILC2, gamma-delta T-cells, regulatory T-cells, NKT-cells, pulmonary-macrophage-resident vs monocyte-derived). The actionable BB use is the foundational data-layer for an AI-DD pipeline designing anti-virulence / anti-persister / anti-biofilm / innate-immune-modulator combinations against drug-resistant Pseudomonas. The first deliverable should be a 2-page Pseudomonas scRNA-seq-atlas handoff brief with foundation-model pretraining data-substrate recommendation (scVI / scGPT-style cross-cohort integration, cell-type-anchor-embedding-learning, infection-trajectory-aware-embedding-prototype), not a compound recommendation. ### 2. Four-axis score \u2014 peptide 0/5; AI-agent infrastructure 5/5 (scRNA-seq time-course atlas is the canonical foundation-layer input for AI-DD target nomination and foundation-model pretraining); longevity 2/5 (acute-infection axis has secondary longevity implications via inflammaging reverse-translation and immune-training biology but is not the primary lane); cost 5/5 (n=5 scRNA-seq atlas is fast to ingest, minimal QC burden, raw MTX/TSV enable immediate CellRanger->Seurat/Scanpy re-analysis). The signal is actionable from the BB multiomics + AI-DD service stack as a respiratory-innate-immunity reference-atlas brief. Why it matters for BB + action \u2014 Ingest GSE345596 raw MTX/TSV files via CellRanger, run standard scRNA-seq QC + integration pipeline (Seurat v4 / Scanpy, Harmony / scVI batch-correction, Leiden-clustering at resolution 0.5-1.5, cell-type-annotation via canonical-marker-panels + AUCell / Garnett classifiers), compute infection-timepoint-trajectory analysis via PAGA + pseudotime (Slingshot / CellRank), and identify the cell-state-transitions that drive Pseudomonas-clearance vs persistence (key-pathway-enrichment via fgsea / decoupleR for IL-17-, TNF-, IL-1beta-, Type-I-IFN-, neutrophil-degranulation-pathways). Package as a candidate list of host-directed-therapeutic targets. ### 3. Signal 2 \u2014 Longevity + Cardiology + Cardiac-Fibrosis + Cross-Species Anti-Remodeling / Adipocyte Enhancer Binding Protein 1 (AEBP1) inhibition mediates cardiac reverse remodeling by regulating MRTFB and RUNX2 signaling [AEBP1-MRTFB-RUNX2 cardiac-fibrosis reverse-remodeling cross-species atlas]: GSE343290 (watcher score 6, n=24, Mus musculus + Homo sapiens cross-species, mixed CUT&Tag/ChIP-seq + RNA-seq, pdat 2026/09/01 = 2 days ago, raw BW/NARROWPEAK files, collected_at 2026-09-02T21:05:54Z) is the FRESHEST NOVEL AEBP1 / MRTFB / RUNX2 / cardiac-fibrosis reverse-remodeling cross-species transcriptomic + chromatin atlas in the entire 2026-09-03 hits.jsonl with the LARGEST-n (n=24) cross-species score-6 NOVEL cohort today. It is a direct heart-failure + cardiac-fibrosis + anti-fibrotic + cross-species-translation + AI-DD target-nomination input for the BB longevity + clinical lane, distinct from yesterday's GSE330687 PD-L1/NSCLC metabolic-reprogramming atlas by tissue (heart vs lung) and from id:87 by mechanism (AEBP1-MRTFB-RUNX2 fibrotic-reverse-remodeling vs PD-L1 metabolic-reprogramming). AEBP1 (Adipocyte-Enhancer-Binding-Protein-1, also called Aortic-Carboxypeptidase-like-Protein, ACLP) is a secreted extracellular-matrix-associated transcriptional-regulator highly-expressed in activated-fibroblasts (myofibroblasts) and stromal-cells with documented roles in (a) cardiac-fibrosis (AEBP1-knockout shows reduced cardiac-fibrosis in pressure-overload [TAC] mouse models via Smad3-collagen-IV-axis attenuation), (b) wound-healing and scar-formation, (c) adipose-tissue-remodeling and metabolic-dysfunction, (d) atherosclerosis and vascular-calcification, (e) tumor-stromal-interactions. MRTFB (Myocardin-Related-Transcription-Factor-B, also called MKL2/MAL16) is a serum-response-factor (SRF) co-activator that drives alpha-SMA/ACTA2, SM22alpha/TAGLN, calponin/CNN1, and myocardin expression in myofibroblasts; RUNX2 (Runt-Related-Transcription-Factor-2) is the master-regulator of osteoblast-differentiation and ectopic-calcification (also implicated in myofibroblast-phenoconversion to osteoblast-like-cells in cardiovascular-calcification). The combined AEBP1-MRTFB-RUNX2 axis thus maps the myofibroblast->osteogenic-fibroblast transition implicated in cardiac-fibrosis + dystrophic-calcification of atherosclerotic-plaques. The actionable BB use is a heart-failure + cardiac-fibrosis + anti-fibrotic + cross-species-translation brief pairing AEBP1-targeting (small-molecule AEBP1-inhibitors in late-preclinical-stage, AEBP1-blocking-antibody program at undisclosed biotechs) with established anti-fibrotic backbone (nintedanib [Ofev, approved for IPF, SSc-ILD, progressive-fibrosing-ILD], pirfenidone [Esbriet, approved for IPF]). ### 4. Signal 3 \u2014 Longevity + Neurodegeneration + Spatial Biology / Spatial transcriptomics of early-grade Huntington's disease and control Caudate [early-HD Caudate spatial atlas]: GSE337131 (watcher score 6, n=2, Homo sapiens, spatial 'Other' [10x Visium HD + Xenium-compatible spatial output, raw CLOUPE/H5AD/JSON/RDS/TIFF files], pdat 2026/09/01 = 2 days ago, raw files yes, collected_at 2026-09-02T21:04:55Z) is the FRESHEST NOVEL early-grade-Huntington's-disease / Caudate-nucleus / spatial-transcriptomic atlas in the entire 2026-09-03 hits.jsonl and the FIRST NOVEL HD spatial-cohort atlas since the score-9 score-band was dedicated to non-neurodegeneration topics. It is a direct neurodegeneration + CAG-repeat-expansion + spatial-cohort atlas + polyQ-protein-aggregation input for the BB longevity + bioinformatics lane. Huntington's-disease (HD) is an autosomal-dominant progressive-neurodegenerative-disorder caused by CAG-trinucleotide-repeat expansion (n>=40, normal n<27) in exon-1 of the HTT-gene encoding the huntingtin-protein (350 kDa, 3144 amino-acids); the expanded polyglutamine (polyQ) tract drives huntingtin-misfolding, -oligomerization, -aggregation (intranuclear-inclusions), and -toxicity (preferentially-striatal-medium-spiny-neuron [MSN] loss -> chorea, dystonia, cognitive-decline, psychiatric-symptoms; ~5-10 per 100,000 prevalence in Western populations, ~3-5 per 100,000 in East-Asian populations, mean-onset 30-50 years with anticipation in paternal transmission). The Caudate-nucleus (a key-striatal-structure [putamen + caudate] with ~50-65% MSN-loss at Vonsattel-grade-3/4 postmortem) is the most-vulnerable-region for HD-pathology and is the primary-imaging-biomarker-target (caudate-atrophy on MRI is a ~10-year-preclinical-phenoconversion-predictor via the TRACK-HD / PREDICT-HD cohorts; ~5-10% caudate-volume-loss/year in premanifest-HD vs ~0.5-1% in age-matched-controls). The actionable BB use is a HTT-lowering + neuroinflammation-modulator + MSN-protectant AI-DD target nomination brief. The first deliverable should be a 2-page HD-spatial-cohort handoff brief with foundation-model pretraining data-substrate recommendation, not a clinical recommendation. The sample-size-2 caveat should be flagged for cross-validation in larger HD-cohort spatial-atlases (the Hu et al. 2024 HD-DREAM challenge dataset, the ENCODE-HiC striatum subset) before client delivery. ### 5. Tracker note \u2014 Biotech Infrastructure / \u00a720 silent-failure fix stable; HF intel layer at 36 events: The 2026-09-03 research-watcher scan returned 105 fresh hits across 27 queries (collected_at 2026-09-02T21:04:27Z, source_errors empty, GEO-query-fan-out completed cleanly, scan.log 936 bytes with `[brown-biotech] watcher done` footer) after the \u00a720 silent-failure fix has now been stable for 5 consecutive days (per the cron-silent-failure-diagnostics skill \u00a720 \u2014 `bash run.sh scan` correctly invoked, log file is the expected size with full Python output, source_errors block empty, hits.jsonl written, scan.json and scan.md written). The HF intel layer's `output/latest/intel.jsonl` continues to hold 36 vendor-intel entries from the 2026-09-03 00:00 KST pre-run intel refresh (36 events across 4 vendors: boltz [12 events: arxiv=10 + hf_tag=2], snap-stanford [9 events: arxiv=4 + hf_tag=4 + hf_org=1], phylo [8 events: arxiv=7 + hf_tag=1], anthropic-science [7 events: arxiv=5 + hf_tag=2]). The 3 NOVEL signals today come from the GEO-source scan only \u2014 the HF intel layer is RECYCLED from prior runs. The 6 NOVEL accessions surfaced today (score >=4, mammalian, pdat >=2026/09/01, absent from id:79-id:87) span 4 BB lanes: (a) Respiratory Infection + Innate Immunity - GSE345596 Pseudomonas-pneumonia scRNA (selected as Signal 1), (b) Longevity + Cardiac-Fibrosis - GSE343290 AEBP1-MRTFB-RUNX2 cardiac-reverse-remodeling (selected as Signal 2), (c) Longevity + Neurodegeneration - GSE337131 early-HD Caudate spatial (selected as Signal 3), (d) Cancer Immunotherapy + scRNA - GSE304905 TM9SF2-reprograms-Neutrophils-NSCLC (score 4, n=3, Mus musculus, pdat 2026/09/01; reserve for tomorrow if a NSCLC-cancer-immunotherapy signal is needed), (e) Longevity + Inflammation - GSE309230 TNF-alpha-TNFR1-pancreatitis (score 4, n=6, Mus musculus, pdat 2026/09/01; reserve for tomorrow if a pancreatic-inflammation signal is needed), (f) Longevity + Kidney - GSE300886 + GSE301168 kidney-disease scRNA + spatial companion-atlas (score 4/4, n=4/4, Mus musculus, pdat 2026/09/01; companion-atlas pairing per the \u00a7Companion-atlas-pairing-rule; reserve for tomorrow if a kidney-fibrosis signal is needed). The score-9 top-band is again fully recycled (per the score-9-recycling pattern noted in id:78-id:87). --- ## 💼 Next Steps - **[Open GSE345596 in GEO (Pseudomonas-pneumonia lung scRNA-seq time-course, n=5)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE345596)** - **[Open GSE343290 in GEO (AEBP1-MRTFB-RUNX2 cardiac-fibrosis cross-species, n=24)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE343290)** - **[Open GSE337131 in GEO (Huntington's-disease Caudate spatial atlas, n=2)](https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE337131)** - **[Request Pseudomonas-innate-immunity + AEBP1-cardiac-fibrosis + HD-Caudate-spatial brief](https://brownbio.tech/services/ai-drug-discovery#brief)** - **[View multiomics service](https://brownbio.tech/services/multiomics)** - **[View biostatx service](https://brownbio.tech/services/biostatx)** - **[View longevity service](https://brownbio.tech/services/longevity)** --- ## 📡 Provenance - **Source:** research-watcher (PubMed/GEO scan, 27 query families × Brown Biotech pipelines) - **Pipeline:** [`brown-biotech-daily-tech-digest`](https://github.com/ohbryt/brown-biotech-platform) — 06:00 KST cron - **Generated:** 2026-09-03 06:10 KST - **Repo:** `ohbryt/brown-biotech-platform` --- _Auto-published by `brown_biotech_research_digest_publisher.py` · [brownbio.tech](https://brownbio.tech) · Decision-ready research, daily._
Brown Biotech Research Digest — 2026-09-03
PubMed/GEO scan · research-watcher · 06:00 KST