Research Pulse — 2026-10-07 | Brown Biotech

# Research Pulse — 2026-10-07
_Generated 2026-10-07 07:32 KST · AI × Bioresearch Daily synthesis_

Brown Biotech daily research pulse — multi-front synthesis across CRISPR /
gene editing, AI drug discovery, protein AI, single-cell foundation models,
FDA/regulatory, and spatial transcriptomics. Sources: company filings,
peer-reviewed journals, FDA guidance, conference disclosures.

---

### Post 1: 오늘의 전체 흐름

🧬 **AI × Bioresearch Daily — 10월 7일 (수요일) · Industrializing(산업화) Day**

10/07(산업화) → 10/04(가속) → 10/03(수렴) → 10/02(정착) → 10/01(임상 전환) → 9/30(번역) → 9/29(안착) → 9/28(제도화). Accelerating(10/04)된 기술들이 **제조·규모화·실제 임상 환자 적용(Industrializing)으로 이행하는 날** — DeWitt et al.로 CRISPR_SCD001 SCD 자가조혈줄기세포 유전자치료 임상후보 제조공정 확립 + NBSGW mice 59% 이식률 + 89%→22% 병적혈색소 감소 + GLP 전전isman的是什么 + first-in-human Phase I trial 착수( Molecular Therapy 2026), Bashir et al.로 CRISPR-Cas9 종양 및 유전성 질환 임상 전환 포괄적 review — BCL11A-enhancer 편집 TDT/SCD 관해 유지 + in vivo TTRamyloidosis LNP 전달 + CEP290-Leber 선천성 흑암 유사증 AAV subretinal 편집 + CRISPR-T 세포 항종양 활성 개선 + DSB-independent(base/pe) 편집 장점/한계 (Frontiers in Genome Editing 2026), Fakiyesi et al.로 CRISPR 심혈관 질환 적용 — inherited cardiovascular diseases 7개 카테고리 체계적 review + cardiomyocytes 재생 한계 + off-target/DNA복구 복잡성 + 게놱 수정 안전성 문제 + 심장نسي세포 특이적 전달 필요성 (Cureus 2026), Iglesias-López et al.로 EU-US CGT 승인 quality/regulatory 비교 — 14개 승인(US 12개, EU 11개) + 모든 제품孤儿指定 + 80% 이상 expedited pathway + 가장 빈번한 quality objections: manufacturing comparability, potency assay validation, specifications, stability + FDA는 data-driven, EMA는 science-based continuous improvement (Therapeutic Innovation & Regulatory Science 2026), Liu et al.로 Gene-Chronos — frozen Geneformer backbone에 learnable temporal prompt tokens + temporal contrastive objective로 발달 시간 추론 + cross-species generalization 확인 (Briefings in Bioinformatics 2026), Weng et al.로 frozen scFM confidence audit framework — raw MSP는 source-target confidence 불일치로 신뢰 불가 + rank-ordering errors + probability drift 분리 + small labeled target data로 recalibration 가능 + automated acceptance/trade-off 문제 (bioRxiv 2026), Gao et al.로 gastric cancer spatial atlas — lymphocyte-aggregated region(LAR) + tertiary lymphoid structures + naïve T cell abundance와 T cell activation pathways 상관관계 + PD1+CD27+CD8 T cells와 CD70+LAMP3+ dendritic cells 공간적 근접성 + immunotherapy biomarker 제시 (Nature Communications 2026), Singh et al.로 AlphaFold drug discovery 적용 한계 — Protein-Ligand data 필요성 hierarchical framework + frontier model dataset 필요성 + SBDD 정확도 boundary 체계적 분석 (Current Opinion Structural Biology 2026)이 동시에 진행된다.

Industrializing은 Accelerating의 다음 단계다. 오늘 6개 전선은 가속화된 기술들이 **제조 공정 확립·규모화·실제 임상 환자 적용으로 이행하며, 기술의 산업화 가능성이 제조·규제·임상 도입 차원에서 동시 확인되는 산업화의 날**이다.

*출처: [DeWitt CRISPR_SCD001 SCD Molecular Therapy 2026](https://consensus.app/papers/details/75e5f43cae74573eb6c6f9f2869f8a49/), [Bashir CRISPR Clinical Translation Frontiers in Genome Editing 2026](https://consensus.app/papers/details/22a428d13f445ebaa1ca1a0a58597a43/), [Fakiyesi CRISPR Cardiovascular Diseases Cureus 2026](https://consensus.app/papers/details/a52297053cfa5471bfa03ab5f8f01db4/), [Iglesias-López CGT Quality Regulatory Therapeutic Innovation 2026](https://consensus.app/papers/details/49b52f00e1d45031af864f85aeb08f46/), [Liu Gene-Chronos Developmental Time Briefings in Bioinformatics 2026](https://consensus.app/papers/details/e09c863192ab54f4bd0d904537a284d5/), [Weng Frozen scFM Confidence Audit bioRxiv 2026](https://consensus.app/papers/details/530d9bca4dcd5352a22a7358413c9999/), [Gao Gastric Cancer Spatial Atlas Nature Communications 2026](https://consensus.app/papers/details/46f3d71acd2e5b6e8816d6886aa485d9/), [Singh AlphaFold Drug Discovery Limitation Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/)*

*#IndustrializingDay #DeWitt #CRISPR_SCD001 #SCD #MolecularTherapy #Manufacturing #NBSGW #Bashir #FrontiersinGenomeEditing #BCL11A #TDT #SCD #TTRAmyloidosis #LNP #CEP290 #LeberCongenital #Fakiyesi #Cureus #Cardiovascular #InheritedDisease #IglesiasLópez #TherapeuticInnovation #RegulatoryScience #Quality #EMA #FDA #ManufacturingComparability #PotencyAssay #Liu #GeneChronos #Geneformer #TemporalPrompt #DevelopmentalTime #BriefingsInBioinformatics #Weng #FrozenFM #ConfidenceAudit #MSP #ProbabilityDrift #Gao #GastricCancer #LAR #TLS #NatureCommunications #PD1 #CD27 #CD8 #Singh #AlphaFold #ProteinLigand #FrontierModel #Industrializing #ClinicalTranslation #ManufacturingScaleUp #RegulatoryQuality #BiomedicalAI #DailyReport*

---

### Post 2: CRISPR & Gene Editing — CRISPR_SCD001 Manufacturing × CRISPR-Cas9 심혈관 질환 적용 × 임상 전환 종합: Industrializing

🪄 **[CRISPR/Gene Editing] DeWitt et al.로 CRISPR_SCD001 SCD 자가조혈줄기세포 유전자치료 임상후보 제조공정 확립 + NBSGW mice 59% 이식률 + 89%→22% 병적혈색소 감소 + GLP 전전ismanانس) + first-in-human Phase I trial 착수(Molecular Therapy 2026) + Bashir et al.로 CRISPR-Cas9 종양 및 유전성 질환 임상 전환 포괄적 review — BCL11A-enhancer 편집 TDT/SCD 관해 유지 + in vivo TTRamyloidosis LNP + CEP290-Leber AAV subretinal + CRISPR-T 항종양 활성 + DSB-independent 편집 장점/한계(Frontiers in Genome Editing 2026) + Fakiyesi et al.로 CRISPR inherited cardiovascular diseases 적용 — 7개 질환 카테고리 + cardiomyocytes 재생 한계 + 게놱 수정 안전성 + 심장نسي세포 특이적 전달 필요성(Cureus 2026): CRISPR Industrializing(산업화) 동시 확인**

CRISPR gene therapy의 **제조 공정 확립 + 심혈관 질환 적용 확대 + 임상 전환 종합(Industrializing)**이 출처에서 동시에 확인된다[1][2][3].

**CRISPR_SCD001: SCD 자가조혈줄기세포 유전자치료 제조공정 확립[1]**
Magdalena DeWitt et al.(Molecular Therapy 2026)의 연구는 CRISPR_SCD001의 임상후보 제조공정과 전임상 결과를 제시한다. **치료 접근**: 자가 CD34+ 세포에서 CRISPR-Cas9 mediated homology-directed repair로 HBB 병적 allele를 교정. **제조공정**: Plerixafor-mobilized peripheral-blood CD34+ cells에서 Large-scale DP lots로 89±5% CD34+ purity와 84±6% viability 달성. **유전자 교정 효율**: 평균 22±4% alleles 교정, 51±6% allelic disruption. **혈색소 회복**: 편집 적혈구 배양에서 성인혈색소 39±5%, 태아혈색소 40±6% 생산. sickle hemoglobin 89%→22±6%로 감소. **전임상 이식**: NBSGW mice에서 59±25% 이식률 + 편집 빈도 유지(~20% 유전자 교정) + 다クロ널 분포 확인 + 처리 관련 독성 없음. **First-in-human Phase I trial 착수**: 제조공정 reproducibility 확인으로 임상 전환 준비 완료. 핵심: CRISPR_SCD001 제조공정 확립 + 전임상 성적 + Phase I trial 착수 — CRISPR 산업화里程碑.

**CRISPR-Cas9 임상 전환 종합: 종양·유전성 질환 포괄적 Review[2]**
Khushi Bashir et al.(Frontiers in Genome Editing 2026)의 review는 CRISPR-Cas9의 종양 및 유전성 질환 임상 전환 현황을 종합한다[2]. **혈색소병**: BCL11A-enhancer 편집으로 태아혈색소 재활성화 + SCD 환자에서 severe vaso-occlusive crises 대부분 관해 유지 + TDT 환자에서 transfusion independence 달성. **ATTR amyloidosis**: in vivo LNP 전달로 혈중 transthyretin protein 수준 감소. **Leber congenital amaurosis type 10**: CEP290 관련증에서 AAV subretinal 전달로 일부 환자에서 임상적으로 유의미한 시각 기능 개선. **CRISPR-T 세포**: 종양 반응성, 지속성, 저항성 개선 설계. **DSB-independent 기술**: base editing과 prime editing이 DSB 관련 위험을 줄일 수 있으나 각 기술별 편집·전달·독성 위험이 있어 제품별 평가 필요. 핵심: CRISPR-Cas9 임상 전환 포괄적 현황 — 산업화段階 종합.

**CRISPR 유전성 심혈관 질환 적용[3]**
Tolulope Fakiyesi et al.(Cureus 2026)의 review는 CRISPR의 inherited cardiovascular diseases 적용을 분석한다[3]. 7개 질환 카테고리 체계적 review: inherited cardiomyopathies, arrhythmic disorders, vascular diseases, lipid disorders, metabolic cardiac conditions, cardiac developmental disorders, systemic disorders affecting heart. **cardiomyocytes 재생 한계**: 성체 심장 cardiomyocytes의 제한적 재생 능력. **게놱 수정 안전성 문제**: off-target effects와 large-scale chromosomal rearrangements 위험. **DNA 복구 메커니즘 복잡성**: DSB 후 복구 경로(NHEJ, HDR)가 편집 결과에 영향. **심장نسي세포 특이적 전달 필요성**: skeletal muscle과 CNS처럼 심장에 도달하는 효율적 전달 체계 필요. 핵심: CRISPR 심혈관 질환 적용의 도전 과제 — 전달 특이성 산업화 과제.

**10월 4일 Accelerating(Lushington PE first-in-human + Jong PE7 + Yu PE efficiency) vs 오늘의 차별점**: 10/04가 **Lushington Molecular Therapy**으로 PE first-in-human functional restoration을, **Jong BMB Reports**으로 PE7까지 7단계 evolution을, **Yu Advanced Science**으로 PE efficiency enhancement를 제시했다면, 오늘은 **DeWitt Molecular Therapy**으로 **CRISPR_SCD001 제조공정 확립 + Phase I trial 착수**를(제조-임상 산업화), **Bashir Frontiers Genome Editing**으로 **CRISPR-Cas9 임상 전환 포괄적 review**를(산업화 단계 종합), **Fakiyesi Cureus**으로 **CRISPR 심혈관 질환 적용의 전달 특이성 과제**를(새 적용 영역 산업화 과제) 동시에 제시한다. Accelerating(10/04) → 제조공정 확립 + 임상 전환 종합 + 새 영역 적용 과제 Industrializing(10/07).

실무 함의: 유전자 치료 개발팀에서 DeWitt et al.의 CRISPR_SCD001 제조공정 결과를 참조하여 gene therapy manufacturing strategy를 수립해야 한다.

*출처: [DeWitt CRISPR_SCD001 SCD Manufacturing Molecular Therapy 2026](https://consensus.app/papers/details/75e5f43cae74573eb6c6f9f2869f8a49/), [Bashir CRISPR Clinical Translation Frontiers Genome Editing 2026](https://consensus.app/papers/details/22a428d13f445ebaa1ca1a0a58597a43/), [Fakiyesi CRISPR Cardiovascular Diseases Cureus 2026](https://consensus.app/papers/details/a52297053cfa5471bfa03ab5f8f01db4/)*

*#CRISPR #GeneEditing #DeWitt #CRISPR_SCD001 #SCD #MolecularTherapy #Manufacturing #GMP #CD34 #Hemoglobin #Bashir #Frontiers #BCL11A #TDT #VasoOcclusive #TTRAmyloidosis #LNP #CEP290 #Leber #CRISPRT #Fakiyesi #Cureus #Cardiovascular #InheritedDisease #Cardiomyocytes #Delivery #IndustrializingDay #ManufacturingScaleUp #PhaseI #ClinicalTranslation #BiomedicalAI #DailyReport*

---

### Post 3: AI 신약 — AI-Caused Accelerated Manufacturing × AlphaFold Drug Discovery Limitation 계층적 인식: Industrializing

💊 **[AI 신약] Singh et al.로 AlphaFold drug discovery 적용 한계 계층적 framework — Protein-Ligand data 필요성 + frontier model dataset 필요성 + SBDD 정확도 boundary 체계적 분석(Current Opinion Structural Biology 2026) + Mao et al.로 AI drug discovery 임상 전환 현황 — Phase IIa efficacy + Phase I safety stop + 0건 완전 AI 승인 + algorithmic transparency + data quality (Frontiers in Pharmacology 2026) + Seyhan et al.로 AI molecule generation + 3D molecular design + AlphaFold/MoleR/PocketCrafter 적용 + single-cell FM + LLM + 0건 완전 마케팅 승인 현실(Critical Reviews Oncology Hematology 2026): AI drug discovery Industrializing(산업화) 동시 확인**

AI-Driven Drug Discovery의 **Protein-Ligand data gap 계층적 인식 + 임상 전환 현황 종합(Industrializing)**이 출처에서 동시에 확인된다[4][5][6].

**AlphaFold Drug Discovery Limitation: Hierarchical Framework + Frontier Model Dataset[4]**
Sukrit Singh et al.(Current Opinion in Structural Biology 2026)의 review는 AlphaFold의 drug discovery 적용 한계를 계층적 framework로 체계화한다[4]. Structure-based drug design(SBDD)이 단백질 구조 정보를 소분자 치료제 설계에 활용하나, AlphaFold 같은 PSPM의 limitations을 blinded challenges로 평가하고 real-world drug design tasks를 반영한 dataset 확장이 필수적이다. **hierarchical framework**: current models가 잘 수행하는 tasks와 challenging/unexplored tasks를 계층적으로 분류. **Frontier model 개발을 위한 dataset 필요성**: experimental + physics-based datasets 생성 노력 highlighting. **SBDD 정확도 boundary**: AlphaFold가 drug discovery tasks에서 정확한 구조 예측을 제공하나, protein dynamics와 multiple conformational states 예측에는 한계. 핵심: AlphaFold drug discovery limitation의 hierarchical framework 체계적 분석 — 산업화 단계 한계 인식.

**AI Drug Discovery 임상 전환 현황: 0건 완전 AI 승인 현실[5]**
Zihan Mao et al.(Frontiers in Pharmacology 2026)의 review는 AI drug discovery의 임상 번역 현황을 종합한다[5]. AI候补 Phase IIa에서 encouraging efficacy를 보인 첫 성과와 Phase I에서 safety signals로 정지된候补이 동시에 존재한다. 완전 AI 발견/설계 약물의 마케팅 승인은 0건이다. AI 독성 earlier 단계 detection이 가능하나 완전히 elimination은 불가하며, biological complexity, patchy data, ML+pharmaceutics 인력 부족이 원인이다. algorithmic transparency와 data quality가 규제 수용성 조건으로 작용한다. 핵심: AI drug discovery 임상 번역 현황 — 0건 완전 AI 승인 현실.

**AI Molecule Generation 종합: AlphaFold/MoleR/PocketCrafter + 0건 마케팅 승인[6]**
A. Seyhan et al.(Critical Reviews in Oncology/Hematology 2026)의 review는 AI molecule generation의 현재 상태를 종합한다[6]. scaffold-aware와 3D molecular design tools(AlphaFold, MoleR, PocketCrafter), single-cell foundation models, large language models(LLM)가 적용된다. target identification, drug discovery, lead optimization, phenotypic screening, precision biology 전 단계에 적용. 그러나 완전 AI 발견/설계 약물의 마케팅 승인은 0건. **향후 방향**: high-quality multimodal data, robust regulatory/ethical frameworks, methodological limitations 인식 필요. 핵심: AI molecule generation 종합 + 0건 완전 마케팅 승인 — 산업화 단계 현실 확인.

**10월 4일 Accelerating(Singh AlphaFold limitation + Yoo TNIK Phase IIa + Lin pharma AI deals) vs 오늘의 차별점**: 10/04가 **Singh Current Opinion**으로 AlphaFold limitation hierarchical framework를, **Yoo BJC Reports**으로 TNIK Phase IIa efficacy를, **Lin GEN Biotechnology**으로 pharma AI deals를 제시했다면, 오늘은 **Singh 재확인**으로 **hierarchical framework + frontier model dataset 필요성**이라는 구체적 산업화 과제를(한계 체계화), **Mao Frontiers Pharmacology**으로 **Phase IIa + Phase I 동시 현실 + 0건 완전 AI 승인**을(임상 산업화 현실), **Seyhan Critical Reviews**으로 **AlphaFold/MoleR/PocketCrafter 적용 + 0건 마케팅 승인**을(분자 생성 산업화 현실) 동시에 제시한다. Accelerating(10/04) → limitation 체계화 + 임상 산업화 현실 + 분자 생성 산업화 현실 Industrializing(10/07).

실무 함의: AI drug discovery 협업팀에서 Singh et al.의 hierarchical framework를 참조하여 AlphaFold 적용 영역 선별 전략을 수립해야 한다.

*출처: [Singh AlphaFold Drug Discovery Limitation Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Mao AI Drug Discovery Clinical Translation Frontiers in Pharmacology 2026](https://consensus.app/papers/details/ca5321af0bb656e095b6108766231764/), [Seyhan AI Drug Discovery Critical Reviews Oncology Hematology 2026](https://consensus.app/papers/details/0e6b685fd2be56fb8a814115873a73d5/)*

*#AI #DrugDiscovery #Singh #AlphaFold #CurrentOpinion #StructuralBiology #ProteinLigand #FrontierModel #HierarchicalFramework #Mao #FrontiersinPharmacology #PhaseIIa #PhaseI #ZeroApproval #Seyhan #CriticalReviews #MolecularDesign #AlphaFold #MoleR #PocketCrafter #LLM #SingleCellFM #IndustrializingDay #ManufacturingLimitations #ClinicalTranslation #BiomedicalAI #DailyReport*

---

### Post 4: 단백질 AI — EU-US CGT Quality 비교 × Manufacturing Comparability Bottleneck: Industrializing

🔬 **[단백질 AI] Iglesias-López et al.로 EU-US CGT 승인 quality/regulatory 비교 — 14개 승인(US 12개, EU 11개) + 모든 제품 orphan designation + 80%+ expedited pathway + 가장 빈번한 quality objections: manufacturing comparability, potency assay validation, specifications, stability + FDA data-driven, EMA science-based continuous improvement(Therapeutic Innovation & Regulatory Science 2026) + Singh et al.로 AlphaFold drug discovery limitation hierarchical framework — Protein-Ligand data 필요성 + frontier model dataset (Current Opinion Structural Biology 2026) + Chakraborty et al.로 AF1→AF2→AF3 architectural evolution + SBDD 가속화 + protein dynamics 한계(Nature Biotechnology 2026): 단백질 AI Industrializing(산업화) 동시 확인**

단백질 AI/규제의 **CGT 제조 quality 비교 체계 + AlphaFold limitation 산업화 과제(Industrializing)**가 출처에서 동시에 확인된다[7][4][8].

**EU-US CGT Quality/Regulatory 비교: Manufacturing Comparability Bottleneck[7]**
C. Iglesias-López et al.(Therapeutic Innovation & Regulatory Science 2026)의 연구는 EU-US CGT 승인의 quality와 regulatory 관점을 비교한다[7]. **승인 현황**: 14개 CGT 승인(US 12개, EU 11개). **Orphan designation**: 모든 제품 orphan designation. **Expedited pathway**: 80% 이상 expedited pathway 활용. **Quality objections 3대 영역**: (1) manufacturing comparability — 제조 공정 변경 시 제품 동등성 입증 어려움, (2) potency assay validation — 효능 측정 분석 방법 검증 표준화 미흡, (3) specifications and stability —产品规格와 안정성 데이터 불충분. **규제 접근 차이**: FDA는 data-driven approach, EMA는 science-based view and continuous improvement. **규제 지원 메커니즘**: expedited submissions 가속화이나 dossiers 질적 개선으로 직접 번역되지 않음. 핵심: EU-US CGT quality 비교 — manufacturing comparability가 主要 bottleneck으로 확립 — 산업화 규제 체계.

**AlphaFold Drug Discovery Limitation: Industrializing 과제[4]**
(Singh Current Opinion Structural Biology 2026 참조 — Post 3에서 상세 기술). 핵심: AlphaFold drug discovery limitation의 hierarchical framework 체계적 분석 — 산업화 단계 한계 인식.

**AlphaFold3 Evolution: AF1→AF2→AF3 Architectural + SBDD[8]**
Chakraborty, Bhattacharya & Lee(Nature Biotechnology 2026)의 review는 AlphaFold 1→2→3의 architectural evolution과 translational research impact를 종합한다[8]. AF1은 DNN, AF2는 Evoformer, AF3는 Pairformer 아키텍처. 2024 Nobel Prize로 protein structure prediction 분야 인정. SBDD 가속화되었으나 protein dynamics와 multiple conformational states 예측 도전과제 남음. 핵심: AF1→AF2→AF3 evolution + SBDD 가속화 — 산업화 architecture.

**10월 4일 Accelerating(Singh AlphaFold limitation + Chakraborty AF3 + Zhang peptide) vs 오늘의 차별점**: 10/04가 **Singh Current Opinion**으로 AlphaFold limitation hierarchical framework를, **Chakraborty Frontiers AI**으로 AF3 evolution을, **Zhang Biotechnology Advances**으로 peptide prediction + MD를 제시했다면, 오늘은 **Iglesias-López Therapeutic Innovation**으로 **EU-US CGT quality 비교에서 manufacturing comparability bottleneck**을(산업화 규제 과제), **Singh 재확인**으로 **hierarchical framework + frontier model dataset 필요성**을(산업화 한계 체계화), 같은 **Chakraborty**에서 **Nature Biotechnology 출처로 구체성**을(architecture evolution 산업화) 동시에 제시한다. Accelerating(10/04) → manufacturing bottleneck 인식 + limitation 체계화 + architecture evolution Industrializing(10/07).

실무 함의: 규제 기획팀에서 Iglesias-López et al.의 EU-US quality 비교 결과를 참조하여 CGT regulatory strategy를 수립해야 한다.

*출처: [Iglesias-López CGT Quality Regulatory Therapeutic Innovation Regulatory Science 2026](https://consensus.app/papers/details/49b52f00e1d45031af864f85aeb08f46/), [Singh AlphaFold Drug Discovery Limitation Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Chakraborty AlphaFold3 Evolution Nature Biotechnology 2026](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/)*

*#ProteinAI #IglesiasLópez #TherapeuticInnovation #RegulatoryScience #CGT #Quality #ManufacturingComparability #PotencyAssay #FDA #EMA #OrphanDesignation #ExpeditedPathway #Singh #AlphaFold #CurrentOpinion #StructuralBiology #HierarchicalFramework #FrontierModel #Chakraborty #AlphaFold3 #NatureBiotechnology #Evolution #Architectural #SBDD #NobelPrize #IndustrializingDay #ManufacturingBottleneck #RegulatoryQuality #BiomedicalAI #DailyReport*

---

### Post 5: Single-Cell Multi-Omics Foundation Model — Gene-Chronos Temporal Inference × Frozen scFM Confidence Audit: Industrializing

🬬 **[Single-Cell Multi-Omics Foundation Model] Liu et al.로 Gene-Chronos — frozen Geneformer backbone에 learnable temporal prompt tokens + temporal contrastive objective로 발달 시간 추론 + cross-species generalization + attention-based developmental genes 식별(Briefings in Bioinformatics 2026) + Weng et al.로 frozen scFM confidence audit framework — raw MSP는 source-target confidence 불일치로 신뢰 불가 + rank-ordering errors + probability drift 분리 + small labeled target data로 recalibration 가능 + automated acceptance/trade-off (bioRxiv 2026) + DenAdel et al.로 scFM plateau — 22.2M cells 400 models 6,400 experiments + scaling law 부재 확인(Nature Methods 2026): foundation model Industrializing(산업화) 동시 확인**

Single-Cell Foundation Model의 **temporal inference 산업화 + confidence reliability 산업화(Industrializing)**가 출처에서 동시에 확인된다[9][10][11].

**Gene-Chronos: Developmental Time Inference from Frozen Geneformer[9]**
Yin-Bo Liu et al.(Briefings in Bioinformatics 2026)의 연구는 Gene-Chronos를 제시한다[9]. **문제**: 기존 temporal inference 방법이 heterogeneous datasets에서 제한된 robustness 보임; scFM은 continuous temporal relationships 포착에 설계되지 않음. **Gene-Chronos 혁신**: frozen pretrained Geneformer backbone에 learnable temporal prompt tokens를 추가. **Temporal contrastive objective**: time-informative signals 추출 + temporally coherent cell representations 조직화. **Cross-species generalization**: multiple benchmark datasets에서 기존 방법 상회 + unseen samples에 strong generalization. **Attention-based 분석**: 발달 진행과 관련된 gene들 식별 → temporal gene expression dynamics에 대한 해석 가능한 통찰 제공. 핵심: Gene-Chronos로 frozen Geneformer 기반 발달 시간 추론 산업화 — scFM의 실용적 적용 확장.

**Frozen scFM Confidence Audit: Source-Target Reliability Gap[10]**
Guang-Zheng Weng et al.(bioRxiv 2026)의 연구는 frozen scFM representations의 confidence estimates를 체계적으로 감사한다[10]. **문제**: scFM predictions가 often confidence assessment 없이 활용되거나, maximum softmax probability(MSP) 단순 cutoff에 의존. **핵심 발견**: raw MSP는 unseen target dataset에서 진정한 confidence를 대표하지 못함. **Source-target confidence gap 원인**: (1) rank-ordering errors, (2) systemic probability drift. **Recalibration 전략**: small labeled target data로 drift successfully recalibration 가능. **Trade-off**: recalibration이 automated acceptance volume을 개선하나 manual review 필요한 cells volume을 증가시킴. **Safe deployment 조건**: representative local labels로 target-specific confidence adjustment 필수. 핵심: frozen scFM confidence audit framework 확립 — 산업화 단계 신뢰성 체계.

**scFM Plateau: Scaling Law 부재 확인[11]**
Alan DenAdel et al.(Nature Methods 2026)의 연구는 scFM의 dataset size/diversity와 performance 관계를 재확인한다[11]. 22.2M cells corpus에서 400개 모델 6,400개 실험 결과, current methods가 일부 크기의 training data에서 plateau 도달. LLM과 달리 명확한 scaling law 관찰되지 않음. 데이터 크기보다 model capacity, dataset size, computational resources의 균형이 중요. 핵심: scFM plateau + scaling law 부재 — 산업화 단계 성능 한계 인식.

**10월 4일 Accelerating(Han scFM benchmark + Pedrocchi SAE + Yan bigger≠better) vs 오늘의 차별점**: 10/04가 **Han bioRxiv**으로 scVI 전체 1위 benchmark를, **Pedrocchi bioRxiv**으로 SAE interpretability를, **Yan BMC Genomics**으로 bigger≠better를 제시했다면, 오늘은 **Liu Briefings in Bioinformatics**으로 **Gene-Chronos — frozen Geneformer에 temporal prompt tokens로 발달 시간 추론**을(temporal inference 산업화), **Weng bioRxiv**으로 **frozen scFM confidence audit — MSP 불신뢰 + rank-ordering errors + recalibration**을(confidence 산업화 과제), **DenAdel Nature Methods**으로 **scaling law plateau 재확인**을(산업화 한계 인식) 동시에 제시한다. Accelerating(10/04) → temporal inference 산업화 + confidence 산업화 과제 + 성능 plateau Industrializing(10/07).

실무 함의: 단일세포 연구팀에서 Weng et al.의 confidence audit framework를 참조하여 scFM deployment reliability strategy를 수립해야 한다.

*출처: [Liu Gene-Chronos Developmental Time Inference Briefings in Bioinformatics 2026](https://consensus.app/papers/details/e09c863192ab54f4bd0d904537a284d5/), [Weng Frozen scFM Confidence Audit bioRxiv 2026](https://consensus.app/papers/details/530d9bca4dcd5352a22a7358413c9999/), [DenAdel scFM Plateau Nature Methods 2026](https://consensus.app/papers/details/f0f6bddc16985563ad8478918950e0f4/)*

*#SingleCell #FoundationModel #Liu #GeneChronos #Geneformer #TemporalPrompt #DevelopmentalTime #BriefingsInBioinformatics #CrossSpecies #Weng #FrozenFM #ConfidenceAudit #MSP #ProbabilityDrift #RankOrdering #Recalibration #DenAdel #NatureMethods #Plateau #NoScalingLaw #IndustrializingDay #TemporalInference #ConfidenceReliability #FoundationModel #BiomedicalAI #DailyReport*

---

### Post 6: FDA/규제 — EU-US CGT Quality Manufacturing Comparability Bottleneck × Manufacturing Scale-Up: Industrializing

📋 **[FDA/규제] Iglesias-López et al.로 EU-US CGT 승인 quality/regulatory 비교 — 14개 승인(US 12개, EU 11개) + 모든 제품 orphan designation + 80%+ expedited + 가장 빈번한 quality objections: manufacturing comparability, potency assay validation, specifications, stability + FDA data-driven, EMA science-based (Therapeutic Innovation & Regulatory Science 2026) + Oo et al.로 FDA platform-aligned + AI/ML analytics — dose selection·safety evaluation·durability prediction convergence (J Clinical Pharmacology 2026) + Shahzad et al.로 FDA CGT 38개 approval assessment — 86.8% orphan + 92.1% expedited + 44.7% surrogate endpoints (Clinical Pharmacology Therapeutics 2026): FDA/규제 Industrializing(산업화) 동시 확인**

FDA 규제 영역의 **EU-US CGT manufacturing quality bottleneck + AI/ML regulatory integration(Industrializing)**이 출처에서 동시에 확인된다[7][12][13].

**EU-US CGT Quality/Regulatory 비교: Manufacturing Comparability Bottleneck[7]**
(Iglesias-López et al. — Post 4에서 상세 기술). 핵심: EU-US CGT quality 비교에서 manufacturing comparability가 主要 bottleneck으로 확립 — 제조 품질 규제 산업화 과제.

**FDA Platform-Aligned AI/ML Analytics Convergence[12]**
C. Oo et al.(J Clinical Pharmacology 2026)의 연구는 FDA gene therapy approvals의 evolvolution과 regulatory framework를 분석한다[12]. FDA의 platform-aligned, risk-based initiatives — plausible mechanism framework, CMC flexibility initiative, advanced manufacturing technologies program. AI/ML-enabled analytics가 dose selection, safety evaluation, durability prediction에서 적용 확대. individualized, mechanistically targeted, potentially curative products 지원을 위한 regulatory framework 발전. 핵심: FDA platform-aligned + AI/ML analytics convergence — 규제-기술 통합 산업화.

**FDA CGT 38개 Approval Assessment: 86.8% Orphan + 92.1% Expedited[13]**
Mahnum Shahzad et al.(Clinical Pharmacology & Therapeutics 2026)의 연구는 FDA CGT 38개 승인의 evidence profile을 분석한다[13]. 86.8% orphan designation, 92.1% expedited pathway, 73.7% single pivotal study, 44.7% surrogate endpoints exclusively, 73.4% PMR/PMC. 17.9%만 primary clinical efficacy endpoint 포함. 핵심: FDA CGT approval assessment 구체적 수치 — 규제 평가 산업화.

**10월 4일 Accelerating(Sheikh Kresladi March 2026 + Oo platform + Shahzad approval) vs 오늘의 차별점**: 10/04가 **Sheikh Annals**으로 Kresladi March 2026 승인을, **Oo J Clinical Pharmacology**으로 platform-aligned를, **Shahzad Clinical Pharmacology**으로 approval assessment 수치를 제시했다면, 오늘은 **Iglesias-López Therapeutic Innovation**으로 **EU-US quality 비교에서 manufacturing comparability bottleneck**이라는 새로운 산업화 관점(manufacturing quality bottleneck)을, **Oo 재확인**으로 **AI/ML analytics dose selection·safety·durability convergence**를(규제-기술 통합 산업화), **Shahzad 재확인**으로 **86.8% orphan + 92.1% expedited** 수치 체계를(규제 평가 산업화) 동시에 제시한다. Accelerating(10/04) → manufacturing quality bottleneck + AI/ML regulatory convergence + approval statistics Industrializing(10/07).

실무 함의: Regulatory Affairs팀에서 Iglesias-López et al.의 EU-US quality 비교 결과를 참조하여 CGT manufacturing quality strategy를 수립해야 한다.

*출처: [Iglesias-López CGT Quality Regulatory Therapeutic Innovation Regulatory Science 2026](https://consensus.app/papers/details/49b52f00e1d45031af864f85aeb08f46/), [Oo FDA Platform AI Analytics J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/), [Shahzad FDA CGT Assessment Clinical Pharmacology Therapeutics 2026](https://consensus.app/papers/details/24705d5037e85bcb9d9863f1aabca040/)*

*#FDA #Regulation #IglesiasLópez #TherapeuticInnovation #RegulatoryScience #CGT #Quality #ManufacturingComparability #PotencyAssay #FDA #EMA #OrphanDesignation #ExpeditedPathway #Oo #JClinicalPharmacology #PlatformAligned #AI #ML #Analytics #DoseSelection #SafetyEvaluation #DurabilityPrediction #Shahzad #ClinicalPharmacology #SurrogateEndpoints #IndustrializingDay #ManufacturingQuality #RegulatoryConvergence #BiomedicalAI #DailyReport*

---

### Post 7: Spatial Transcriptomics — Gastric Cancer LAR Atlas × Lymphocyte-Aggregated Region × Immunotherapy Biomarker: Industrializing

🗺️ **[Spatial Transcriptomics] Gao et al.로 gastric cancer spatial atlas — lymphocyte-aggregated region(LAR) + tertiary lymphoid structures + naïve T cell abundance와 T cell activation pathways 상관관계 + PD1+CD27+CD8 T cells와 CD70+LAMP3+ dendritic cells 공간적 근접성 + immunotherapy biomarker 제시(Nature Communications 2026) + Li et al.로 pan-cancer 56 LCPs + Niche_4(macrophage+tumor) poor prognosis + Niche_11(macrophage+immune) better survival (Cell Reports Medicine 2026) + Zhang et al.로 SCLC CosMx 600K 세포 + PIHs-1 survival predictor + lymph node metastasis (Cell Reports Medicine 2026): spatial transcriptomics Industrializing(산업화) 동시 확인**

Spatial transcriptomics의 **gastric cancer LAR atlas 확립 + immunotherapy biomarker 제시(Industrializing)**가 출처에서 동시에 확인된다[14][15][16].

**Gastric Cancer Spatial Atlas: Lymphocyte-Aggregated Region(LAR) + TLS[14]**
Sen Gao et al.(Nature Communications 2026)의 연구는 gastric cancer의 공간적 atlas를 제시한다[14]. **4개 공간 영역**: single-cell + spatial transcriptomics 통합으로 cellular spatial distribution을 분석. **lymphocyte-aggregated region(LAR)**: lymphocyte aggregates와 tertiary lymphoid structures(TLS)로 구성. **Naïve T cell abundance와 T cell activation pathways 상관관계**: naïve T cell abundance ↑와 T cell activation-associated pathways ↑ 관련. **세포 공간적 분포**: Group A(더 활성화된 lymphocyte)와 Group B(더 휴면状态的 lymphocyte)로 분화. **PD1+CD27+CD8 T cells와 CD70+LAMP3+ dendritic cells**: 공간적 근접성 확인. **Immunotherapy biomarker 제시**: PD1+CD27+CD8 T cells와 CD70+LAMP3+ DCs 근접성이 immunotherapy response 예측 biomarke로 제시. 핵심: gastric cancer LAR atlas 확립 + TLS 기반 immunotherapy biomarker — 공간 면역 종양학 산업화.

**Pan-Cancer 56 LCPs + Niche Survival Analysis[15]**
Jiarong Li et al.(Cell Reports Medicine 2026)의 연구는 12개 암종 373개 검체에서 56개 지역 세포 프로그램(LCPs)과 13개 반복 niches를 분석한다[15]. Niche_4(macrophage + tumor cell)는 poor prognosis + immunotherapy resistance 연관. Niche_11(macrophage + immune cell)이 better survival + treatment response 연관. 핵심: pan-cancer niche survival analysis — 산업화 생물학적 validation.

**SCLC 600K Cells + PIHs-1 Survival Predictor[16]**
Zicheng Zhang et al.(Cell Reports Medicine 2026)의 연구는 CosMx SCLC에서 600K 세포를 profiling한다[16]. 3개 LNM-enriched malignant subclusters, vascular-immune crosstalk, PIHs-1 survival predictor 확인. 핵심: SCLC spatial map + PIHs-1 — 임상 적용 산업화.

**10월 4일 Accelerating(Cervilla 5-platform + Li 56 LCPs + Zhang SCLC) vs 오늘의 차별점**: 10/04가 **Cervilla Genome Biology**으로 5개 플랫폼 benchmark를, **Li Cell Reports Medicine**으로 Niche_4/Niche_11 survival predictors를, **Zhang Cell Reports Medicine**으로 PIHs-1을 제시했다면, 오늘은 **Gao Nature Communications**으로 **gastric cancer LAR atlas + TLS + PD1+CD27+CD8/CD70+LAMP3+ DCs 근접성 biomarker**을(새 고형암 적용 + immunotherapy biomarker 산업화), **Li 재확인**으로 **Niche_4/Niche_11**을(산업화 생물학적 validation), **Zhang 재확인**으로 **PIHs-1**을(산업화 생존 예측) 동시에 제시한다. Accelerating(10/04) → gastric cancer LAR atlas + immunotherapy biomarker + niche validation Industrializing(10/07).

실무 함의: 면역종양학팀에서 Gao et al.의 LAR/TLS 분석을 참조하여 gastric cancer immunotherapy patient selection strategy를 수립해야 한다.

*출처: [Gao Gastric Cancer Spatial Atlas Lymphocyte-Aggregated Region Nature Communications 2026](https://consensus.app/papers/details/46f3d71acd2e5b6e8816d6886aa485d9/), [Li Pan-Cancer Spatial Niche Cell Reports Medicine 2026](https://consensus.app/papers/details/f420ab985e1257e18352cac9e0c425cf/), [Zhang CosMx SCLC Cell Reports Medicine 2026](https://consensus.app/papers/details/dc8b02ddbfcb574fb92b18cf8aed5117/)*

*#SpatialTranscriptomics #Gao #GastricCancer #NatureCommunications #LAR #LymphocyteAggregatedRegion #TLS #TertiaryLymphoidStructures #PD1 #CD27 #CD8 #CD70 #LAMP3 #DendriticCells #Immunotherapy #Biomarker #Li #PanCancer #56LCPs #Niche4 #Niche11 #Macrophage #Tumor #Prognosis #Zhang #CosMx #SCLC #PIHs1 #SurvivalPredictor #LymphNodeMetastasis #IndustrializingDay #ImmunotherapyBiomarker #SpatialAtlas #BiomedicalAI #DailyReport*

---

### Post 8: 오늘의 요약 — Industrializing(산업화) Day 핵심 정리

🧬 **오늘 6개 영역 핵심 요약:**

10/07(Industrializing)의 핵심: Accelerating(10/04)된 기술들이 **제조 공정 확립·규모화·실제 임상 환자 적용(Industrializing)으로 이행하며, 기술의 산업화 가능성이 제조·규제·임상 도입 차원에서 동시 확인되는 산업화의 날**이었다.

**1. CRISPR_SCD001 제조공정 확립 + Phase I Trial 착수(DeWitt Molecular Therapy 2026)**: Large-scale DP lots로 89% CD34+ purity + 84% viability + NBSGW mice 59% 이식률 + sickle hemoglobin 89%→22% 감소 + first-in-human Phase I trial 착수 — 제조공정 산업화里程碑[1].

**2. CRISPR-Cas9 임상 전환 포괄적 Review(Bashir Frontiers in Genome Editing 2026)**: BCL11A-enhancer 편집 TDT/SCD 관해 유지 + TTR amyloidosis LNP + CEP290 AAV subretinal + CRISPR-T 항종양 + DSB-independent 편집 장점/한계 — 임상 전환 산업화 종합[2].

**3. EU-US CGT Quality 비교: Manufacturing Comparability Bottleneck(Iglesias-López Therapeutic Innovation 2026)**: US 12개, EU 11개 승인 + 모든 제품 orphan + 80%+ expedited + manufacturing comparability가 主要 bottleneck + FDA data-driven vs EMA science-based — 제조 품질 규제 산업화 과제[7].

**4. Gene-Chronos: Frozen Geneformer 발달 시간 추론(Liu Briefings in Bioinformatics 2026)**: frozen Geneformer backbone + learnable temporal prompt tokens + temporal contrastive objective + cross-species generalization — temporal inference 산업화[9].

**5. Frozen scFM Confidence Audit: MSP 불신뢰(Weng bioRxiv 2026)**: raw MSP source-target 불일치 + rank-ordering errors + probability drift + recalibration 필요 — confidence reliability 산업화 과제[10].

**6. Gastric Cancer LAR Atlas + Immunotherapy Biomarker(Gao Nature Communications 2026)**: lymphocyte-aggregated region + TLS + PD1+CD27+CD8/CD70+LAMP3+ DCs 근접성 biomarker — 고형암 spatial immunotherapy biomarker 산업화[14].

10월 8일 전망: Industrializing된 기술들의 제조 공정 규모화와 실제 임상 환자 적용 속도가 핵심 변수. 특히 CRISPR_SCD001 Phase I trial enrollment 시작, Gao gastric cancer LAR biomarker의 prospective validation, Weng confidence audit framework의 임상 적용 가능성이 관전이다.

*출처: [DeWitt CRISPR_SCD001 SCD Manufacturing Molecular Therapy 2026](https://consensus.app/papers/details/75e5f43cae74573eb6c6f9f2869f8a49/), [Bashir CRISPR Clinical Translation Frontiers Genome Editing 2026](https://consensus.app/papers/details/22a428d13f445ebaa1ca1a0a58597a43/), [Fakiyesi CRISPR Cardiovascular Diseases Cureus 2026](https://consensus.app/papers/details/a52297053cfa5471bfa03ab5f8f01db4/), [Singh AlphaFold Drug Discovery Limitation Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Iglesias-López CGT Quality Regulatory Therapeutic Innovation 2026](https://consensus.app/papers/details/49b52f00e1d45031af864f85aeb08f46/), [Mao AI Drug Discovery Frontiers in Pharmacology 2026](https://consensus.app/papers/details/ca5321af0bb656e095b6108766231764/), [Seyhan AI Drug Discovery Critical Reviews Oncology Hematology 2026](https://consensus.app/papers/details/0e6b685fd2be56fb8a814115873a73d5/), [Chakraborty AlphaFold3 Nature Biotechnology 2026](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/), [Liu Gene-Chronos Briefings in Bioinformatics 2026](https://consensus.app/papers/details/e09c863192ab54f4bd0d904537a284d5/), [Weng Frozen scFM Confidence Audit bioRxiv 2026](https://consensus.app/papers/details/530d9bca4dcd5352a22a7358413c9999/), [DenAdel scFM Plateau Nature Methods 2026](https://consensus.app/papers/details/f0f6bddc16985563ad8478918950e0f4/), [Gao Gastric Cancer Spatial Atlas Nature Communications 2026](https://consensus.app/papers/details/46f3d71acd2e5b6e8816d6886aa485d9/), [Li Pan-Cancer Spatial Niche Cell Reports Medicine 2026](https://consensus.app/papers/details/f420ab985e1257e18352cac9e0c425cf/), [Zhang CosMx SCLC Cell Reports Medicine 2026](https://consensus.app/papers/details/dc8b02ddbfcb574fb92b18cf8aed5117/), [Oo FDA Platform AI J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/), [Shahzad FDA CGT Clinical Pharmacology 2026](https://consensus.app/papers/details/24705d5037e85bcb9d9863f1aabca040/)*

*#IndustrializingDay #Summary #DeWitt #CRISPR_SCD001 #Manufacturing #PhaseI #Bashir #Frontiers #BCL11A #TDT #SCD #TTRAmyloidosis #CEP290 #Fakiyesi #Cardiovascular #IglesiasLópez #Quality #ManufacturingComparability #PotencyAssay #FDA #EMA #Singh #AlphaFold #HierarchicalFramework #FrontierModel #Mao #ZeroApproval #Seyhan #MolecularDesign #Chakraborty #AlphaFold3 #Liu #GeneChronos #Geneformer #TemporalInference #Weng #ConfidenceAudit #MSP #DenAdel #Plateau #Gao #GastricCancer #LAR #TLS #PD1 #CD27 #CD8 #CD70 #LAMP3 #Li #Niche4 #Niche11 #Zhang #PIHs1 #Oo #PlatformAligned #AI #ML #Shahzad #Orphan #Expedited #Industrializing #ManufacturingScaleUp #ClinicalTranslation #RegulatoryQuality #BiomedicalAI #DailyReport*

---

## Sources (1차 출처 10개 이상)

[1] [DeWitt et al. — Development of CRISPR_SCD001, an Autologous Hematopoietic Stem Cell Gene Therapy for Sickle Cell Disease after CRISPR-Cas9 Mediated Correction. Molecular Therapy 2026](https://consensus.app/papers/details/75e5f43cae74573eb6c6f9f2869f8a49/)
[2] [Bashir et al. — Clinical translation of CRISPR-Cas9 therapeutics in cancer and inherited genetic disorders. Frontiers in Genome Editing 2026](https://consensus.app/papers/details/22a428d13f445ebaa1ca1a0a58597a43/)
[3] [Fakiyesi et al. — CRISPR-Based Genome Editing in Inherited Cardiovascular Diseases: Therapeutic Advances, Emerging Editing Platforms, and Translational Challenges. Cureus 2026](https://consensus.app/papers/details/a52297053cfa5471bfa03ab5f8f01db4/)
[4] [Singh et al. — More Protein-Ligand data is needed for AlphaFold-like Models to enable drug discovery. Current Opinion in Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/)
[5] [Mao et al. — Artificial intelligence in drug discovery: from algorithmic foundations to clinical translation. Frontiers in Pharmacology 2026](https://consensus.app/papers/details/ca5321af0bb656e095b6108766231764/)
[6] [Seyhan et al. — AI in Drug Discovery and Development. Critical Reviews in Oncology/Hematology 2026](https://consensus.app/papers/details/0e6b685fd2be56fb8a814115873a73d5/)
[7] [Iglesias-López et al. — Outlook of Cell Gene Therapies Development and Approval from Quality and Regulatory Perspective. Therapeutic Innovation & Regulatory Science 2026](https://consensus.app/papers/details/49b52f00e1d45031af864f85aeb08f46/)
[8] [Chakraborty, Bhattacharya & Lee — The transformative impact of AI-enabled AlphaFold 3: evolution, current status, and future prospects. Nature Biotechnology 2026](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/)
[9] [Liu et al. — Gene-Chronos: parameter-efficient developmental time inference using a pretrained single-cell foundation model. Briefings in Bioinformatics 2026](https://consensus.app/papers/details/e09c863192ab54f4bd0d904537a284d5/)
[10] [Weng et al. — Cross-domain confidence reliability and remappability of frozen single-cell representations. bioRxiv 2026](https://consensus.app/papers/details/530d9bca4dcd5352a22a7358413c9999/)
[11] [DenAdel et al. — Evaluating the role of pre-training dataset size and diversity on single-cell foundation model performance. Nature Methods 2026](https://consensus.app/papers/details/f0f6bddc16985563ad8478918950e0f4/)
[12] [Gao et al. — A spatially resolved atlas of gastric cancer characterises a lymphocyte-aggregated region. Nature Communications 2026](https://consensus.app/papers/details/46f3d71acd2e5b6e8816d6886aa485d9/)
[13] [Li et al. — Pan-cancer analysis of spatial transcriptomics reveals heterogeneous tumor spatial microenvironment. Cell Reports Medicine 2026](https://consensus.app/papers/details/f420ab985e1257e18352cac9e0c425cf/)
[14] [Zhang et al. — Single-cell spatial transcriptomics reveals tumor microenvironment heterogeneity in primary and lymph node-metastatic small cell lung cancer. Cell Reports Medicine 2026](https://consensus.app/papers/details/dc8b02ddbfcb574fb92b18cf8aed5117/)
[15] [Oo et al. — FDA Gene Therapy Approvals (1998–2025): Current Status, Regulatory Evolution, and Future Directions. J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/)
[16] [Shahzad et al. — Assessment of United States Food and Drug Administration Approval of Cell and Gene Therapies. Clinical Pharmacology & Therapeutics 2026](https://consensus.app/papers/details/24705d5037e85bcb9d9863f1aabca040/)