Research Pulse — 2026-10-10 | Brown Biotech

# Research Pulse — 2026-10-10
_Generated 2026-10-10 07:30 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 10일 (금요일) · Leveraging(실현) Day**

10/10(실현) → 10/09(임상 전환) → 10/08(성숙) → 10/07(산업화) → 10/06(산업화) → 10/05(확장) → 10/04(가속) → 10/03(수렴). Translating(10/09)된 기술들이 **실제 임상 데이터·규제 승인을 기반한 구체적 이행(Leveraging)의 단계로 진입하는 날** — Khare et al.로 CRISPR 임상 시험 scoping review: 29건 trials 420명 환자 median 6개월~90세 + CRISPR-Cas9 79% + 白血病·림프종 CR 15-92% + 혈색소질환 지속적 호전(2026 Cytotherapy), Lummerstorfer et al.로 non-viral CRISPR carriers: 136건 trials 중 36건 in vivo + LNP가비iral主流 + transient delivery = off-target 감소 + hit-and-run fashion(Drug Delivery 2026), Oo et al.로 FDA CGT 44건 승인 종합 + platform-aligned risk-based approach + AI/ML dose selection·safety evaluation·durability prediction convergence(J Clinical Pharmacology 2026), Sheikh et al.로 Kresladi(marnetogra-gene autotemcel) LAD-I FDA 승인 2026년 3월 26일: 첫 gene therapy for LAD-I + pediatric 대상(ITGB2 돌연변이)(Annals Medicine Surgery 2026), Singh et al.로 AlphaFold protein-ligand data gap: hierarchical framework + frontier model dataset 필요성(Current Opinion Structural Biology 2026), Ji et al.로 CAPTAIN multimodal: 4M+ cells에서 transcriptomes + 382 proteins 동시 측정 + protein imputation·expansion·cell type annotation·batch harmonization + COVID-19 immune interaction patterns(Nature Communications 2026), Zhang et al.로 SCLC CosMx 600K+ cells: 75명 105개 검체 + PIHs-1 survival predictor + vascular-immune crosstalk(Cell Reports Medicine 2026), Li et al.로 HiST deep learning: histology images에서 spatial GEPs prediction + 5개 암종 AUC 0.96 + Pearson 0.74 + TCGA prognostic stratification(Advanced Science 2026)이 동시에 진행된다.

Leveraging은 Translating의 다음 단계다. 오늘 6개 전선은 임상 전환된 기술들이 **실제 규제 승인·실제 제조 공정 확립·실제 임상 데이터로 구체적 이행(Leveraging)하는 날**이다.

*출처: [Khare CRISPR Clinical Trials Scoping Cytotherapy 2026](https://consensus.app/papers/details/5e9d45a6ee2659ee92d7247ea2339ef8/), [Lummerstorfer Non-Viral CRISPR Carriers Drug Delivery 2026](https://consensus.app/papers/details/747d35e9e5bc5ca9a9854998c7acd326/), [Oo FDA Gene Therapy Approvals J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/), [Sheikh LAD-I Gene Therapy FDA Approval Annals Medicine Surgery 2026](https://consensus.app/papers/details/f2f27f7392e05d9693a11da0cb87b29d/), [Singh AlphaFold Protein-Ligand Data Gap Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Ji CAPTAIN Multimodal Foundation Nature Communications 2026](https://consensus.app/papers/details/40361e55d4815c1dae3d5f6a95d4468a/), [Zhang CosMx SCLC Cell Reports Medicine 2026](https://consensus.app/papers/details/dc8b02ddbfcb574fb92b18cf8aed1117/), [Li HiST Spatial Transcriptomics Advanced Science 2026](https://consensus.app/papers/details/5dfa277e84425924aac9d148317c0bb7/)*

*#LeveragingDay #Khare #Cytotherapy #CRISPR #ClinicalTrials #ScopingReview #29Trials #420Patients #Lummerstorfer #DrugDelivery #NonViral #LNPs #136Trials #HitAndRun #Oo #FDA #CGT #44Approvals #PlatformAligned #AIAnalytics #Sheikh #LAD-I #Kresladi #FDAApproval #March2026 #ITGB2 #Pediatric #Singh #AlphaFold #ProteinLigand #DataGap #CurrentOpinion #Ji #CAPTAIN #NatureCommunications #Multimodal #382Proteins #4MillionCells #Zhang #CosMx #SCLC #PIHs1 #VascularImmune #Li #HiST #AdvancedScience #Histology #SpatialGEPs #AUC #Translating #ClinicalTranslation #RealWorldData #RegulatoryApproval #BiomedicalAI #DailyReport*

---

### Post 2: CRISPR & Gene Editing — 136건 Trials + Non-Viral Shift + Kresladi Approval

🪄 **[CRISPR/Gene Editing] Khare et al.로 CRISPR 임상 trial scoping review: 29건 trials 420명 환자 median 6개월~90세 + CRISPR-Cas9 79% + 白血病/림프종 CR 15-92% + 혈색소질환 지속적 호전 + 장기 안전성 데이터 부족(2026 Cytotherapy) + Lummerstorfer et al.로 non-viral CRISPR carriers: 136건 trials 중 36건 in vivo + LNP가비iral主流 + transient delivery = off-target 감소 + hit-and-run fashion(Drug Delivery 2026) + Sheikh et al.로 Kresladi(marnetogra-gene autotemcel) LAD-I FDA 승인 2026년 3월 26일: 첫 gene therapy for LAD-I + pediatric 대상 + ITGB2 돌연변이 + 75% 2세 사망률 historical + HSCT 75% 생존율 + matched sibling donor 없는 환자에게 적용(Annals Medicine Surgery 2026): CRISPR Leveraging(실현) 동시 확인**

CRISPR gene therapy의 **29건 trial scoping review + 136건 trials 중 36건 in vivo non-viral 현실 + Kresladi FDA 승인(2026)(Leveraging)**이 출처에서 동시에 확인된다[1][2][3].

**CRISPR 임상 Trial Scoping Review: 29건 Trial 420명 Patient 현황[1]**
Gabrielle Khare et al.(Cytotherapy 2026)의 scoping review는 CRISPR gene editing 임상 시험의 현재 모습을 체계적으로 분석한다[1]. **29건 published trials 420명 환자** 분석 — median age 6개월~90세. **CRISPR-Cas9가 79%(23건 trial)**로 절대적主流. **대상 질환**: 白血病·림프종 27.5%(8건) + 혈색소질환 20.6%(6건) + 고형종양 20.6%(6건) + 희귀유전질환 17.2%(5건). **완전 관해(CR)률**: 白血病/림프종 15-92%. **혈색소질환**: 편집 세포 3개월·6개월 시점 검출 + 평균 혈색소 수준 지속적 호전. **Transfection 방식**: electroporation 31% + viral-mediated 20.7% + lipid nanoparticles(LNPs) 10.3%. **In vivo 6건**，其余均为ex vivo. **Phase I/II early-phase 단일 대조 연구 1건**除外. **장기 종양발생 모니터링 데이터 부족**. 핵심: CRISPR 임상 trial 29건 scoping review — 이행(Leveraging) 단계.

**Non-Viral CRISPR Carriers: 136건 Trials 중 36건 In Vivo[2]**
M. Lummerstorfer et al.(Drug Delivery 2026)의 review는 non-viral CRISPR 전달의 임상 시험 현황을 분석한다[2]. **136건 CRISPR trials ongoing(2025년 12월 기준)**. **36건(26%)이 in vivo 전달** — 명확한 비바이러스성 전환. **Non-viral vs viral**: 유전체 편집은 영구적이나 CRISPR machinery 장기 발현 불필요 → transient delivery로 충분. **Hit-and-run fashion**: 짧은 발현 = off-target 위험 감소. **LNP가 in vivo CRISPR의主流 전달체**. **Clinical employed technologies**: Cas9 RNPs, base editing components, prime editing components. 핵심: non-viral CRISPR carriers 136건 trial 동향 — in vivo LNP主流 구현(Leveraging).

**Kresladi LAD-I FDA 승인 2026년 3월 26일: 첫 Gene Therapy for LAD-I[3]**
M. Sheikh et al.(Annals of Medicine and Surgery 2026)의 연구는 Kresladi(marnetogra-gene autotemcel)의 FDA 승인을 보고한다[3]. **Leukocyte adhesion deficiency type I(LAD-I)**: ITGB2 유전자 돌연변이로 白血球가 감염과 싸우지 못함. **2세 사망률 75% historical** + moderate disease 환자의 40세 사망률 50%. **기존 치료**: stem cell transplantation — matched sibling donor 없으면 이식 실패 위험 높음(graft-versus-host disease). **Kresladi 승인 배경**: (1) FDA 2026년 3월 26일 승인, (2) 첫 gene therapy for LAD-I, (3) pediatric patients 대상, (4) matched sibling donor가 없는 환자에게 적용. 핵심: Kresladi FDA 승인 2026년 3월 26일 — Casgevy 이후 새 규제 승인(Leveraging).

**10월 9일 Translating(Şahin 700 trials + Şahin monitoring toolkit + Kunanayagam pediatric) vs 오늘의 차별점**: 10/09가 **Şahin Molecular Therapy Nucleic Acids**으로 700건 global trials 등록 현황을, **Şahin Nature Reviews Genetics**으로 off-target monitoring toolkit을, **Kunanayagam Pediatric Research**으로 14개 pediatric CGT 적용을 제시했다면, 오늘은 **Khare Cytotherapy**으로 **29건 trial scoping review로 420명 환자의 체계적 분석**(이행의 체계적 실현)을, **Lummerstorfer Drug Delivery**으로 **136건 trials 중 36건 in vivo non-viral + LNP主流的具体적 수치**(이행의 제조 실현)를, **Sheikh Annals**으로 **Kresladi FDA 승인 2026년 3월 26일이라는 구체적 날짜와 적응증**(규제 승인의 실현)을 동시에 제시한다. Translating(10/09) → trial scoping review 체계적 실현 + non-viral delivery 제조 실현 + Kresladi 규제 승인 실현 Leveraging(10/10).

실무 함의: 유전자 치료 개발팀에서 Lummerstorfer et al.의 136건 trial 동향 중 36건 in vivo non-viral 현실을 참조하여 delivery strategy를 점검해야 한다.

*출처: [Khare CRISPR Clinical Trials Scoping Review Cytotherapy 2026](https://consensus.app/papers/details/5e9d45a6ee2659ee92d7247ea2339ef8/), [Lummerstorfer Non-Viral CRISPR Carriers Drug Delivery 2026](https://consensus.app/papers/details/747d35e9e5bc5ca9a9854998c7acd326/), [Sheikh LAD-I Gene Therapy Kresladi FDA Approval Annals Medicine Surgery 2026](https://consensus.app/papers/details/f2f27f7392e05d9693a11da0cb87b29d/)*

*#CRISPR #GeneEditing #Khare #Cytotherapy #ScopingReview #29Trials #420Patients #CRISPRCas9 #79Percent #Electroporation #LNPs #10Percent #Lummerstorfer #DrugDelivery #NonViral #InVivo #136Trials #HitAndRun #TransientDelivery #OffTarget #Sheikh #LAD-I #Kresladi #FDAApproval #March2026 #ITGB2 #Pediatric #GeneTherapy #LAD #HSCT #LeveragingDay #ManufacturingRealization #RegulatoryApproval #BiomedicalAI #DailyReport*

---

### Post 3: AI 신약 — Chen 2025 + Ali 2026 + Ferreira 2025: 3-Year Consensus Summary

💊 **[AI 신약] Chen et al.로 AI drug discovery evolution: ML/DL 통합으로 drug characterization·target discovery·small molecule design·clinical trials 가속화 + molecular generation·virtual screening + timeline 60% 단축 가능성(2025 J Pharmaceutical Analysis) + Ali et al.로 AI-Pharma collaborative models: algorithmic transparency·data quality + interoperability + regulatory acceptance 필수 조건 + AI-driven in silico platforms로 toxicity·PK·developability 예측(2026 Drug Development Research) + Ferreira et al.로 AI/ML comprehensive review 2019-2024: deep learning·GNNs·transformers + target identification·lead discovery·preclinical safety assessment + data accessibility·interpretability·clinical translation 문제(2025 ACS Omega): AI Drug Discovery Leveraging(실현) 동시 확인**

AI-Driven Drug Discovery의 **3-Year Consensus(2025-2026) 종합(Leveraging)**이 출처에서 동시에 확인된다[4][5][6].

**AI Drug Discovery Evolution: ML/DL 통합으로 전 영역 가속화[4]**
Chen Fu et al.(J Pharmaceutical Analysis 2025)의 review는 AI의 drug discovery 통합을 종합한다[4]. **ML/DL 통합 영역**: drug characterization + target discovery and validation + small molecule drug design + clinical trials efficiency. **Molecular generation**: novel drug molecules 생성 + property·activity 예측. **Virtual screening**:候补 최적화. **Clinical trial 가속**: outcome 예측 + trial 설계 + drug repositioning. **Timeline 60% 단축 가능성**: 신약 개발 timeline大幅 단축. 핵심: AI drug discovery evolution — 3년 내 통합 가속화(Leveraging).

**AI-Pharma Collaborative Models: 필수 조건[5]**
M. Ali et al.(Drug Development Research 2026)의 review는 AI-pharma 협업 모델을 분석한다[5]. **Algorithmic transparency**: 규제 승인의 전제 조건. **Data quality**: 모델 신뢰성 기반. **Interoperability**: 시스템 상호운용성. **Regulatory acceptance**: 승인을 위한 관료적 요구사항. **AI-driven in silico platforms**: toxicity prediction + pharmacokinetics(PK) prediction + developability assessment → late-stage attrition 감소. 핵심: AI-Pharma collaborative models 필수 조건 — 이행의 협업 체계(Leveraging).

**AI/ML Comprehensive Review 2019-2024: Deep Learning·GNNs·Transformers[6]**
Fábio J. N. Ferreira et al.(ACS Omega 2025)의 review는 2019-2024년 AI/ML drug discovery를 종합한다[6]. **Methods**: deep learning + graph neural networks(GNNs) + transformers. **적용 영역**: target identification + lead discovery + lead optimization + preclinical safety assessment. **Persistent gaps**: data accessibility + model interpretability + clinical translation. **Ethical frameworks 필요성**. 핵심: AI/ML 5년 review — 이행의 성과와 과제(Leveraging).

**10월 9일 Translating(Seyhan Phase I/II + Mao Phase IIa + Ali collaborative) vs 오늘의 차별점**: 10/09가 **Seyhan Critical Reviews**으로 Phase I/II候补 동시 존재를, **Mao Frontiers Pharmacology**으로 Phase IIa efficacy + Phase I safety stop를, **Ali Expert Opinion**으로 algorithmic transparency·data quality regulatory acceptance를 제시했다면, 오늘은 **Chen J Pharmaceutical Analysis 2025**으로 **ML/DL 통합으로 전 영역 가속화 + timeline 60% 단축 가능성**이라는 통합적 관점(3년 Consensus Leveraging)을, **Ali Drug Development Research 2026 재확인**으로 **algorithmic transparency·data quality + AI-driven in silico platforms로 toxicity·PK·developability 예측**이라는 협업 체계(협업 Leveraging)를, **Ferreira ACS Omega 2025**으로 **2019-2024 5개년 review로 deep learning·GNNs·transformers의 구체적 적용 영역과 persistent gaps**를(성과와 과제 Leveraging) 동시에 제시한다. Translating(10/09) → 3년 consensus 통합 + 협업 체계 + 성과와 과제 Leveraging(10/10).

실무 함의: AI drug discovery 협업팀에서 Chen 2025의 timeline 60% 단축 가능성을 참조하여 development strategy를 점검해야 한다.

*출처: [Chen AI Pharmaceuticals J Pharmaceutical Analysis 2025](https://consensus.app/papers/details/1308c134b9165903b7d81c4b8c636084/), [Ali AI Drug Development Drug Development Research 2026](https://consensus.app/papers/details/7a8e73e032a7505fa67758810d3c03ec/), [Ferreira AI Drug Discovery ACS Omega 2025](https://consensus.app/papers/details/4cde73a603a4581da45ee8c2734fb28a/)*

*#AI #DrugDiscovery #Chen #JPharmaceuticalAnalysis #2025 #Ali #DrugDevelopmentResearch #2026 #Ferreira #ACSOmega #2025 #MachineLearning #DeepLearning #GNNs #Transformers #Timeline #60Percent #AlgorithmicTransparency #DataQuality #Interoperability #RegulatoryAcceptance #InSilico #Toxicity #PK #Developability #MolecularGeneration #VirtualScreening #LeadDiscovery #TargetIdentification #LeveragingDay #3YearConsensus #CollaborationModels #BiomedicalAI #DailyReport*

---

### Post 4: 단백질 AI — AlphaFold Protein-Ligand Data Gap Hierarchical Framework

🔬 **[단백질 AI] Singh et al.로 AlphaFold protein-ligand data gap hierarchical framework: protein-ligand complex data 필요성 + blinded challenges로 평가 필요 + frontier model dataset 필요성 + SBDD 정확도 boundary 체계적 분석(Current Opinion Structural Biology 2026) + Fang et al.로 AlphaFold3: Google DeepMind + Isomorphic Labs 개발 + unprecedented accuracy + protein·ligand·nucleic acid·ions 모두 예측 + drug design·vaccine development·precision therapy 적용(Precision Clinical Medicine 2025) + Chakraborty et al.로 AF1→AF2→AF3 evolution: Evoformer→Pairformer + SBDD 가속화 + protein dynamics 한계(Nature Biotechnology 2026): 단백질 AI Leveraging(실현) 동시 확인**

단백질 AI의 **AlphaFold drug discovery 적용의 데이터 갭 계층화(Leveraging)**가 출처에서 동시에 확인된다[7][8][9].

**AlphaFold Protein-Ligand Data Gap: Hierarchical Framework[7]**
Sukrit Singh et al.(Current Opinion in Structural Biology 2026)의 review는 AlphaFold의 drug discovery 적용 한계를 계층적 framework로 체계화한다[7]. **Structure-based drug design(SBDD)**: 단백질 구조 정보를 소분자 치료제 설계에 활용. **Protein-ligand complex data 필요성**: AlphaFold는 protein structure는 정확하나 protein-ligand complex는 별도 데이터 필요. **Hierarchical framework 필요성**: current models가 잘 수행하는 tasks vs challenging/unexplored tasks를 계층적으로 분류. **Frontier model 개발을 위한 systematic dataset generation 필요성**: experimental + physics-based datasets 생성 노력 highlighting. **SBDD 정확도 boundary**: protein dynamics + multiple conformational states 예측에 한계. 핵심: AlphaFold protein-ligand data gap hierarchical framework — 이행의 데이터 기반 한계 인식(Leveraging).

**AlphaFold3: Google DeepMind + Isomorphic Labs 개발[8]**
Ziqi Fang et al.(Precision Clinical Medicine 2025)의 review는 AlphaFold3의 특성을 분석한다[8]. **개발**: Google DeepMind + Isomorphic Labs 공동 개발. **Unprecedented accuracy**: virtually 모든 biomolecule 구조 예측. **예측 범위**: proteins + ligands + nucleic acids + ions + modified residues. **적용**: drug design + vaccine development + precision therapy. **AF3 vs AF2 차이**: Evoformer(Pairformer로 교체)으로 pairwise amino acid interactions 모델링. 핵심: AlphaFold3 drug discovery 적용의 기술적 기반 — 이행의 기술 실현(Leveraging).

**AlphaFold Evolution: AF1→AF2→AF3 Architecture[9]**
Chakraborty, Bhattacharya & Lee(Nature Biotechnology 2026)의 review를 재확인한다[9]. AF1(DNN) → AF2(Evoformer) → AF3(Pairformer). 2024 Nobel Prize로 protein structure prediction 인정. SBDD 가속화되었으나 protein dynamics + multiple conformational states 예측 도전과제 남음. 핵심: AlphaFold evolution + SBDD 가속화 — 이행의 architecture 실현(Leveraging).

**10월 9일 Translating(Feldman AF3 adversarial + ESMFold 우위) vs 오늘의 차별점**: 10/09가 **Feldman Computational Structural Biotechnology**으로 AF3 adversarial mutation + 40% residues 불변 + confidence metrics 35%만 정확을 제시했다면, 오늘은 **Singh Current Opinion Structural Biology**으로 **protein-ligand complex data gap의 hierarchical framework 체계적 분석**이라는 이행의 데이터 기반 한계 인식(Leveraging)을, **Fang Precision Clinical Medicine**으로 **AlphaFold3의 기술적 기반(unprecedented accuracy + drug design/vaccine/precision therapy 적용)**이라는 이행의 기술 실현(Leveraging)을, **Chakraborty 재확인**으로 **architecture evolution + SBDD 가속화**를(architecture Leveraging) 동시에 제시한다. Translating(10/09) → 데이터 갭 계층화 + 기술 기반 + architecture evolution Leveraging(10/10).

실무 함의: SBDD 워크플로우팀에서 Singh et al.의 hierarchical framework를 참조하여 AlphaFold 적용 영역과 한계를 선별적으로 인식해야 한다.

*출처: [Singh AlphaFold Protein-Ligand Data Gap Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Fang AlphaFold3 Precision Clinical Medicine 2025](https://consensus.app/papers/details/ed02d9aca58c5d7c9bdec1a2b9f9cb13/), [Chakraborty AlphaFold3 Nature Biotechnology 2026](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/)*

*#ProteinAI #AlphaFold #Singh #CurrentOpinion #StructuralBiology #2026 #ProteinLigand #DataGap #HierarchicalFramework #FrontierModel #Dataset #SBDD #Fang #PrecisionClinicalMedicine #2025 #AlphaFold3 #DeepMind #IsomorphicLabs #Unprecedented #Accuracy #DrugDesign #Vaccine #PrecisionTherapy #Chakraborty #NatureBiotechnology #Evoformer #Pairformer #SBDD #NobelPrize #ProteinDynamics #LeveragingDay #DataGap #TechnologyRealization #Architecture #BiomedicalAI #DailyReport*

---

### Post 5: Single-Cell Multi-Omics Foundation Model — CAPTAIN Multimodal Clinical Utility

🬬 **[Single-Cell Multi-Omics Foundation Model] Ji et al.로 CAPTAIN multimodal foundation model: 4M+ cells에서 transcriptomes + 382 proteins 동시 측정 + protein imputation·expansion·cell type annotation·batch harmonization + COVID-19 immune interaction patterns + unified multimodal representations(Nature Communications 2026) + Wang et al.로 SCMBench 23 methods: FMs < DMs in multi-omics integration + lightweight adaptation strategy로 FM 성능 향상(Nature Communications 2026) + Baek et al.로 single-cell foundation models review: transformer architecture + self-supervised learning + scFMs challenges + non-sequential data nature + data quality inconsistency + computational intensity(Experimental Molecular Medicine 2025): foundation model Leveraging(실현) 동시 확인**

Single-Cell Foundation Model의 **CAPTAIN multimodal clinical utility 실증(Leveraging)**이 출처에서 동시에 확인된다[10][11][12].

**CAPTAIN: Multimodal Transcriptome + Protein Foundation Model[10]**
Bo-Ya Ji et al.(Nature Communications 2026)의 연구는 CAPTAIN을 제시한다[10]. **4M+ cells에서 동시 측정**: transcriptomes + 382 proteins concurrently measured. **기존 scFMs 한계**: transcriptrime-only 훈련 → biased and partial cellular characterizations. **CAPTAIN 혁신**: unified multimodal representations by modelling cross-modality dependencies + capturing cellular state diversity. **Protein imputation**: 전사체에서 단백질 수준 예측. **Protein expansion**: 기존 데이터에서 단백질 보강. **Cell type annotation**: 세포 유형 주석 개선. **Batch harmonization**: 배치 효과 보정. **Zero-shot generalization**: unseen tasks에서 강인한 일반화. **COVID-19 immune interaction patterns**: 단백질-유전체 통합으로才发现 — protein-driven intercellular dynamics. 핵심: CAPTAIN multimodal (transcriptome + 382 proteins) foundation model — 임상 적용 가능성 실증(Leveraging).

**SCMBench: FMs < DMs in Multi-Omics Integration[11]**
Yixuan Wang et al.(Nature Communications 2026)의 SCMBench를 재확인한다[11]. **23개 방법 평가**: FMs < DMs in multi-omics integration. **Lightweight adaptation strategy**: FM 성능 향상 가능. 핵심: SCMBench benchmarking 결과 — 이행의 성능 격차(Leveraging).

**Single-Cell Foundation Models Review: Challenges[12]**
S. Baek et al.(Experimental & Molecular Medicine 2025)의 review를 재확인한다[12]. **Transformer architecture + self-supervised learning**: scFMs 핵심. **Challenges**: non-sequential data nature + data quality inconsistency + computational intensity. **Biological relevance**: latent embeddings의 해석 어려움. 핵심: scFMs challenges — 이행의 계산적 제약(Leveraging).

**10월 9일 Translating(Ji CAPTAIN + Wang SCMBench + Cen MedComm) vs 오늘의 차별점**: 10/09가 **Ji Nature Communications**으로 CAPTAIN multimodal을, **Wang Nature Communications**으로 SCMBench FMs < DMs를, **Cen MedComm**으로 generative AI + foundation models를 제시했다면, 오늘은 **Ji 재확인**으로 **CAPTAIN의 4M+ cells + 382 proteins 동시 측정에서 protein imputation/expansion/cell type annotation/batch harmonization의 구체적 성능과 COVID-19 immune interaction patterns 발견**이라는 임상 적용 가능성 실증(Leveraging)을, **Wang 재확인**으로 **FMs < DMs benchmarking 현실**을(성능 격차 Leveraging), **Baek Experimental Molecular Medicine**으로 **scFMs challenges — non-sequential data + data quality + computational intensity**를(계산적 제약 Leveraging) 동시에 제시한다. Translating(10/09) → CAPTAIN clinical utility 실증 + benchmarking 현실 + 계산적 제약 Leveraging(10/10).

실무 함의: 단일세포 연구팀에서 Ji et al.의 CAPTAIN multimodal 접근법을 proteome 통합 analysis strategy에 적용해야 한다.

*출처: [Ji CAPTAIN Multimodal Foundation Model Nature Communications 2026](https://consensus.app/papers/details/40361e55d4815c1dae3d5f6a95d4468a/), [Wang SCMBench Nature Communications 2026](https://consensus.app/papers/details/24b273d01c1f508c9a20a5a981ac2e3c/), [Baek Single-Cell Foundation Models Experimental Molecular Medicine 2025](https://consensus.app/papers/details/e3168426724b5c3c8d867083d35fdc92/)*

*#SingleCell #FoundationModel #Ji #CAPTAIN #NatureCommunications #2026 #Multimodal #Transcriptome #Protein #382Proteins #4MillionCells #Imputation #Expansion #CellTypeAnnotation #BatchHarmonization #COVID19 #ImmuneInteraction #Wang #SCMBench #NatureCommunications #FMsLessThanDMs #LightweightAdaptation #23Methods #Baek #ExperimentalMolecularMedicine #2025 #Transformer #SelfSupervised #Challenges #NonSequential #DataQuality #ComputationalIntensity #LeveragingDay #ClinicalUtility #BiomedicalAI #DailyReport*

---

### Post 6: FDA/규제 — FDA CGT 44건 Approval + Platform-Aligned + AI/ML Analytics

📋 **[FDA/규제] Oo et al.로 FDA Gene Therapy Approvals 1998-2025: 44건 CGT 승인(2025년 5월 기준) + platform-aligned risk-based approach + AI/ML dose selection·safety evaluation·durability prediction convergence(J Clinical Pharmacology 2026) + Shahzad et al.로 FDA CGT 38건 approval assessment: 86.8% orphan + 92.1% expedited + 73.7% single pivotal study + 44.7% surrogate endpoints + 73.4% PMR/PMC(Clinical Pharmacology Therapeutics 2026) + Sheikh et al.로 Kresladi FDA 승인 2026년 3월 26일: 첫 gene therapy for LAD-I + ITGB2 + pediatric 대상(Annals Medicine Surgery 2026): FDA/규제 Leveraging(실현) 동시 확인**

FDA 규제 영역의 **44건 CGT 승인 현실 + AI/ML analytics convergence + Kresladi FDA 승인(Leveraging)**이 출처에서 동시에 확인된다[3][13][14].

**FDA Gene Therapy Approvals 44건: Platform-Aligned AI/ML Convergence[13]**
C. Oo et al.(J Clinical Pharmacology 2026)의 review를 재확인한다[13]. **44건 CGT 승인(2025년 5월 기준)**: gene therapy의 본격적 임상 적용 확인. **Platform-aligned, risk-based approach**: plausible mechanism framework + CMC flexibility initiative + advanced manufacturing technologies program. **AI/ML analytics**: dose selection + safety evaluation + durability prediction convergence. **Individualized, mechanistically targeted, potentially curative products** 지원. 핵심: FDA 44건 CGT approvals + AI/ML analytics convergence — 규제-기술 통합 이행(Leveraging).

**FDA CGT 38건 Approval Assessment: 86.8% Orphan + 92.1% Expedited[14]**
Mahnum Shahzad et al.(Clinical Pharmacology & Therapeutics 2026)을 재확인한다[14]. **86.8% orphan designation** + **92.1% expedited pathway** + **73.7% single pivotal study** + **44.7% surrogate endpoints exclusively** + **73.4% PMR/PMC**. 핵심: FDA CGT 구체적 승인 수치 체계 — 규제 평가 이행(Leveraging).

**Kresladi FDA 승인 2026년 3월 26일[3]**
(M. Sheikh et al. — Post 2에서 상세 기술). 핵심: Kresladi FDA 승인 2026년 3월 26일 — Casgevy 이후 새 규제 승인(Leveraging).

**10월 9일 Translating(Oo 44 CGT + Shahzad approval statistics + Kunanayagam pediatric) vs 오늘의 차별점**: 10/09가 **Oo J Clinical Pharmacology**으로 44건 CGT approvals를, **Shahzad Clinical Pharmacology**으로 approval statistics를, **Kunanayagam Pediatric Research**으로 14개 pediatric CGT를 제시했다면, 오늘은 **Oo 재확인**으로 **44건 CGT approvals + platform-aligned AI/ML convergence**의 구체적 이행(Leveraging)을, **Shahzad 재확인**으로 **86.8% orphan + 92.1% expedited + 73.7% single study + 44.7% surrogate endpoints** 수치 체계(Leveraging)를, **Sheikh 재확인**으로 **Kresladi FDA 승인 2026년 3월 26일이라는 구체적 날짜**를(규제 승인의 실현 Leveraging) 동시에 제시한다. Translating(10/09) → 44건 approvals 이행 + statistics 체계 + Kresladi 구체적 승인 Leveraging(10/10).

실무 함의: Regulatory Affairs팀에서 Oo et al.의 44건 CGT 승인 현황을 gene therapy regulatory strategy에 반영해야 한다.

*출처: [Oo FDA Gene Therapy Approvals J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/), [Shahzad FDA CGT Assessment Clinical Pharmacology Therapeutics 2026](https://consensus.app/papers/details/24705d5037e85bcb9d9863f1aabca040/), [Sheikh LAD-I Kresladi FDA Approval Annals Medicine Surgery 2026](https://consensus.app/papers/details/f2f27f7392e05d9693a11da0cb87b29d/)*

*#FDA #GeneTherapy #Oo #JClinicalPharmacology #44CGTApprovals #PlatformAligned #RiskBased #AIAnalytics #DoseSelection #SafetyEvaluation #DurabilityPrediction #Shahzad #ClinicalPharmacology #Therapeutics #86Percent #Orphan #92Percent #Expedited #73Percent #SingleStudy #44Percent #SurrogateEndpoints #73Percent #PMR #PMC #Sheikh #Kresladi #LAD-I #FDAApproval #March2026 #ITGB2 #Pediatric #GeneTherapy #LeveragingDay #RegulatoryApproval #BiomedicalAI #DailyReport*

---

### Post 7: Spatial Transcriptomics & Pan-Cancer — Pan-Cancer Niche + SCLC CosMx + HiST

🗺️ **[Spatial Transcriptomics] Li et al.로 pan-cancer spatial transcriptomics: 12개 암종 373개 검체 + 56개 LCPs + 13개 niches + Niche_4(macrophage+tumor) poor prognosis + immunotherapy resistance + Niche_11(macrophage+immune) better survival + treatment response(Cell Reports Medicine 2026) + Zhang et al.로 SCLC CosMx: 75명 105개 검체 600K+ cells + PIHs-1 survival predictor + vascular-immune crosstalk + LNM-enriched malignant subclusters 3개(Cell Reports Medicine 2026) + Li et al.로 HiST deep learning: histology images에서 spatial GEPs prediction + 5개 암종 AUC 0.96 + Pearson 0.74 + TCGA prognostic stratification(Advanced Science 2026): spatial transcriptomics Leveraging(실현) 동시 확인**

Spatial transcriptomics의 **pan-cancer niche differential + SCLC CosMx + HiST reconstruction(Leveraging)**이 출처에서 동시에 확인된다[15][16][17].

**Pan-Cancer Spatial Niche: Niche_4 vs Niche_11 Differential[15]**
Jiarong Li et al.(Cell Reports Medicine 2026)의 연구를 재확인한다[15]. **12개 암종 373개 검체**에서 **56개 LCPs + 13개 niches** 식별. **Niche_4(macrophage + tumor cell)**: poor prognosis + immunotherapy resistance 연관. **Niche_11(macrophage + immune cell)**: better survival + treatment response 연관. **공간적 조직**: 종양 세포와 대식세포의 위치 관계가 환자 결과 예측. 핵심: pan-cancer spatial niche differential — 이행의 biomarker 실현(Leveraging).

**SCLC CosMx 600K+ Cells: PIHs-1 + Vascular-Immune Crosstalk[16]**
Zi-Cheng Zhang et al.(Cell Reports Medicine 2026)의 연구를 재확인한다[16]. **75명 환자 105개 검체**: 600K+ cells profiling. **3개 LNM-enriched malignant subclusters**: 림프절 전이 강화 종양 하위 집합. **PIHs-1**: 독립적 survival predictor. **Vascular-immune crosstalk**: endothelial cells가 immune activation orchestrate + cytotoxic T cells와 functional perivascular niches 형성. 핵심: SCLC CosMx spatial atlas + PIHs-1 — 이행의 공간적 biomarker 실현(Leveraging).

**HiST: Deep Learning Spatial Transcriptomics Reconstruction[17]**
Wei Li et al.(Advanced Science 2026)의 연구를 재확인한다[17]. **Histology images에서 spatial GEPs prediction**: 조직 슬라이드 이미지만으로 공간적 유전자 발현 프로파일 예측. **5개 암종 AUC 0.96**: breast cancer 포함 다수 암종에서 높은 정확도. **Pearson correlation 0.74**: 예측된 GEPs와 실제 ST 데이터 상관관계. **TCGA prognostic stratification**: 예측된 GEPs로 환자 예후分层. **Implications**: 종양 이질성 평가 + 면역치료 반응 예측 + 생물학적 경로 분석. 핵심: HiST deep learning spatial transcriptomics reconstruction — 이행의 계산적 biomarker 실현(Leveraging).

**Pan-Cancer vs SCLC Differential**: Pan-cancer niche(Li)는 Niche_4/Niche_11의 보편적 패턴을, SCLC CosMx(Zhang)는 폐소세포폐암 특이적 LNM enriched clusters와 PIHs-1 predictor를, HiST(Li)는 조직학 이미지만으로 spatial GEPs를 prediction하는 계산적 접근을 제시한다. 동일 biomarker 영역이지만 보편적 패턴 vs 암종特異적 atlas vs 계산적 reconstruction으로 차별화된다.

**10월 9일 Translating(Zhang SCLC + Gao LAR + Li niche) vs 오늘의 차별점**: 10/09가 **Zhang Cell Reports Medicine**으로 SCLC CosMx 600K+ cells + PIHs-1를, **Gao Nature Communications**으로 gastric cancer LAR atlas를, **Li Cell Reports Medicine**으로 Niche_4/Niche_11를 제시했다면, 오늘은 **Li 재확인**으로 **12개 암종 373개 검체 56 LCPs 13 niches의 보편적 Niche_4/Niche_11 differential**이라는 보편적 biomarker 체계(Leveraging)를, **Zhang 재확인**으로 **SCLC CosMx 75명 105개 검체 600K+ cells의 구체적 수치와 LNM enriched subclusters**를(암종特異적 Leveraging), **Li HiST 재확인**으로 **histology images에서 AUC 0.96 + Pearson 0.74의 계산적 biomarker reconstruction**을(계산적 Leveraging) 동시에 제시한다. Translating(10/09) → pan-cancer niche differential + SCLC CosMx 구체적 수치 + HiST 계산적 biomarker Leveraging(10/10).

실무 함의: 면역종양학팀에서 Li et al.의 Niche_4/Niche_11 differential을 immunotherapy patient selection biomarker strategy에 적용해야 한다.

*출처: [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/dc8b02ddbfcb574fb92b18cf8aed1117/), [Li HiST Spatial Transcriptomics Advanced Science 2026](https://consensus.app/papers/details/5dfa277e84425924aac9d148317c0bb7/)*

*#SpatialTranscriptomics #Li #PanCancer #CellReportsMedicine #56LCPs #13Niches #Niche4 #Niche11 #Macrophage #Tumor #Prognosis #ImmunotherapyResistance #Zhang #CosMx #SCLC #75Patients #105Specimens #600KCells #PIHs1 #VascularImmune #Crosstalk #LNMMalignant #Subclusters #Li #HiST #AdvancedScience #Histology #SpatialGEPs #AUC #Pearson #TCGA #PrognosticStratification #LeveragingDay #BiomarkerRealization #BiomedicalAI #DailyReport*

---

### Post 8: 오늘의 요약 — Leveraging(실현) Day 핵심 정리

🧬 **오늘 6개 영역 핵심 요약:**

10/10(Leveraging)의 핵심: Translating(10/09)된 기술들이 **실제 규제 승인·실제 제조 공정 확립·실제 임상 데이터로 구체적 이행(Leveraging)하는 날**이었다.

**1. CRISPR Clinical Trial 29건 Scoping + Non-Viral 136건 Trials(Khare Cytotherapy 2026 + Lummerstorfer Drug Delivery 2026)**: 29건 trial 420명 patient에서 Cas9 79% + CR 15-92% + 혈색소질환 지속적 호전; 136건 trials 중 36건 in vivo non-viral + LNP主流 + hit-and-run = off-target 감소 — 이행의 임상 데이터 실현[1][2].

**2. Kresladi FDA 승인 2026년 3월 26일: LAD-I Gene Therapy(Sheikh Annals Medicine Surgery 2026)**: FDA 2026년 3월 26일 승인 + 첫 gene therapy for LAD-I + ITGB2 돌연변이 + pediatric 대상 + matched sibling donor 없는 환자에게 적용 — 규제 승인의 실현[3].

**3. AI Drug Discovery 3-Year Consensus: Chen 2025 + Ali 2026 + Ferreira 2025**: ML/DL 통합으로 전 영역 가속화 + timeline 60% 단축 가능성 + algorithmic transparency·data quality·interoperability·regulatory acceptance 필수 조건 — 이행의 협업 체계 실현[4][5][6].

**4. AlphaFold Protein-Ligand Data Gap Hierarchical Framework(Singh Current Opinion Structural Biology 2026)**: protein-ligand complex data 필요성 + hierarchical framework로 current vs challenging tasks 분류 + frontier model dataset 필요성 — 이행의 데이터 기반 한계 인식[7].

**5. CAPTAIN Multimodal (Transcriptome + 382 Proteins) Clinical Utility(Ji Nature Communications 2026)**: 4M+ cells에서 transcriptomes + 382 proteins 동시 측정 + protein imputation/expansion/cell type annotation/batch harmonization + COVID-19 immune interaction patterns — 임상 적용 가능성 실증[10].

**6. Pan-Cancer Niche + SCLC CosMx + HiST Biomarker Realization(Li Cell Reports Medicine 2026 + Zhang Cell Reports Medicine 2026 + Li Advanced Science 2026)**: Niche_4/Niche_11 differential(12개 암종 373개 검체) + SCLC CosMx 600K+ cells PIHs-1 + HiST histology→spatial GEPs AUC 0.96 — biomarker 이행의 실현[15][16][17].

Leveraging은 Translating의 다음 단계다. 오늘의 6개 전선은 그 순자의 열여섯 번째 단계에 있다.

내일(10월 11일 토요일) 바라볼 세 지점: Kresladi 승인 이후 LAD-I 치료제 확대 가능성, CAPTAIN multimodal의 추가 조직 적용, HiST의 임상 pathology 적용 속도가 관전이다.

Leveraging 다음은 ** Consolidating(집약)**이다. 오늘의 6개 전선이 제시하는 공통 질문: 이행된 기술들이 각 분야主流 체계에서 어떤 방식으로 집약(Consolidating)되어야 하는가. 이 질문이 내일 이후 각 전선의 핵심 과제가 된다.

*출처: [Khare CRISPR Clinical Trials Scoping Cytotherapy 2026](https://consensus.app/papers/details/5e9d45a6ee2659ee92d7247ea2339ef8/), [Lummerstorfer Non-Viral CRISPR Carriers Drug Delivery 2026](https://consensus.app/papers/details/747d35e9e5bc5ca9a9854998c7acd326/), [Sheikh LAD-I Gene Therapy Kresladi FDA Approval Annals Medicine Surgery 2026](https://consensus.app/papers/details/f2f27f7392e05d9693a11da0cb87b29d/), [Chen AI Pharmaceuticals J Pharmaceutical Analysis 2025](https://consensus.app/papers/details/1308c134b9165903b7d81c4b8c636084/), [Ali AI Drug Development Drug Development Research 2026](https://consensus.app/papers/details/7a8e73e032a7505fa67758810d3c03ec/), [Ferreira AI Drug Discovery ACS Omega 2025](https://consensus.app/papers/details/4cde73a603a4581da45ee8c2734fb28a/), [Singh AlphaFold Protein-Ligand Data Gap Current Opinion Structural Biology 2026](https://consensus.app/papers/details/81a3128ae1615cfeb1bd6544d3ad2f1b/), [Fang AlphaFold3 Precision Clinical Medicine 2025](https://consensus.app/papers/details/ed02d9aca58c5d7c9bdec1a2b9f9cb13/), [Chakraborty AlphaFold3 Nature Biotechnology 2026](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/), [Ji CAPTAIN Multimodal Foundation Nature Communications 2026](https://consensus.app/papers/details/40361e55d4815c1dae3d5f6a95d4468a/), [Wang SCMBench Nature Communications 2026](https://consensus.app/papers/details/24b273d01c1f508c9a20a5a981ac2e3c/), [Baek Single-Cell Foundation Models Experimental Molecular Medicine 2025](https://consensus.app/papers/details/e3168426724b5c3c8d867083d35fdc92/), [Oo FDA Gene Therapy Approvals J Clinical Pharmacology 2026](https://consensus.app/papers/details/181658e93c825ea581a0664845a98d12/), [Shahzad FDA CGT Assessment Clinical Pharmacology Therapeutics 2026](https://consensus.app/papers/details/24705d5037e85bcb9d9863f1aabca040/), [Li Pan-Cancer Niche Cell Reports Medicine 2026](https://consensus.app/papers/details/f420ab985e1257e18352cac9e0c425cf/), [Zhang CosMx SCLC Cell Reports Medicine 2026](https://consensus.app/papers/details/dc8b02ddbfcb574fb92b18cf8aed1117/), [Li HiST Advanced Science 2026](https://consensus.app/papers/details/5dfa277e84425924aac9d148317c0bb7/)*

*#LeveragingDay #Summary #Khare #Cytotherapy #29Trials #420Patients #Lummerstorfer #DrugDelivery #136Trials #NonViral #Sheikh #Kresladi #LAD-I #FDAApproval #March2026 #Chen #JPharmaceuticalAnalysis #Ali #DrugDevelopmentResearch #Ferreira #ACSOmega #Singh #AlphaFold #ProteinLigand #DataGap #Ji #CAPTAIN #382Proteins #Wang #SCMBench #Baek #Oo #44CGTApprovals #Shahzad #Orphan #Expedited #Li #PanCancer #56LCPs #Niche4 #Niche11 #Zhang #CosMx #PIHs1 #Li #HiST #AUC #Leveraging #ClinicalTranslation #RegulatoryApproval #ManufacturingRealization #BiomedicalAI #DailyReport*

---

## Sources (1차 출처 17개 이상)

[1] [Khare G, et al. A scoping review of gene editing in clinical trials. Cytotherapy. 2026.](https://consensus.app/papers/details/5e9d45a6ee2659ee92d7247ea2339ef8/)
[2] [Lummerstorfer M, et al. Non-Viral CRISPR carriers: transient delivery with lasting effects. Drug Delivery. 2026.](https://consensus.app/papers/details/747d35e9e5bc5ca9a9854998c7acd326/)
[3] [Sheikh M, et al. Gene therapy for leukocyte adhesion deficiency type I: FDA approval, clinical evidence, and future perspectives. Annals of Medicine and Surgery. 2026.](https://consensus.app/papers/details/f2f27f7392e05d9693a11da0cb87b29d/)
[4] [Chen F, et al. The future of pharmaceuticals: Artificial intelligence in drug discovery and development. J Pharmaceutical Analysis. 2025.](https://consensus.app/papers/details/1308c134b9165903b7d81c4b8c636084/)
[5] [Ali M, et al. Artificial Intelligence in Drug Discovery and Development: Transforming Pharmaceutical Innovation. Drug Development Research. 2026.](https://consensus.app/papers/details/7a8e73e032a7505fa67758810d3c03ec/)
[6] [Ferreira FJN, et al. AI-Driven Drug Discovery: A Comprehensive Review. ACS Omega. 2025.](https://consensus.app/papers/details/4cde73a603a4581da45ee8c2734fb28a/)
[7] [Singh S, 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/)
[8] [Fang Z, et al. AlphaFold 3: an unprecedented opportunity for fundamental research and drug development. Precision Clinical Medicine. 2025.](https://consensus.app/papers/details/ed02d9aca58c5d7c9bdec1a2b9f9cb13/)
[9] [Chakraborty C, Bhattacharya S, Lee J. The transformative impact of AI-enabled AlphaFold 3. Nature Biotechnology. 2026.](https://consensus.app/papers/details/34db0d4f32255b8696144613f6182df8/)
[10] [Ji BY, et al. CAPTAIN: a multimodal foundation model pretrained on co-assayed single-cell RNA and protein. Nature Communications. 2026.](https://consensus.app/papers/details/40361e55d4815c1dae3d5f6a95d4468a/)
[11] [Wang Y, et al. SCMBench: benchmarking domain-specific and foundation models for single-cell multi-omics data integration. Nature Communications. 2026.](https://consensus.app/papers/details/24b273d01c1f508c9a20a5a981ac2e3c/)
[12] [Baek S, et al. Single-cell foundation models: bringing artificial intelligence into cell biology. Experimental & Molecular Medicine. 2025.](https://consensus.app/papers/details/e3168426724b5c3c8d867083d35fdc92/)
[13] [Oo C, 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/)
[14] [Shahzad M, 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/)
[15] [Li J, et al. Pan-cancer analysis of spatial transcriptomics reveals heterogeneous tumor spatial microenvironment. Cell Reports Medicine. 2026.](https://consensus.app/papers/details/f420ab985e1257e18352cac9e0c425cf/)
[16] [Zhang ZC, 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/dc8b02ddbfcb574fb92b18cf8aed1117/)
[17] [Li W, et al. HiST: Histological Images Reconstruct Tumor Spatial Transcriptomics via MultiScale Fusion Deep Learning. Advanced Science. 2026.](https://consensus.app/papers/details/5dfa277e84425924aac9d148317c0bb7/)