{"slug": "ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone", "title": "AI-designed neoantigen vaccines just hit a Phase 3 milestone —", "summary": "AI-designed neoantigen vaccines have reached a Phase 3 milestone, with the pipeline from tumor sequencing to mRNA construct increasingly driven by machine learning tools such as NetMHCpan, MHCflurry, and pVACseq. The author outlines a reproducible end-to-end stack for GPU clusters, noting that prediction accuracy is no longer the bottleneck—NetMHCpan-4.1 achieves AUC 0.92—but manufacturing turnaround remains slow, with Moderna quoting 6-8 weeks from biopsy to vial and academic cores taking 12-16 weeks.", "body_md": "# AI-designed neoantigen vaccines just hit a Phase 3 milestone —\n\n*work*at scale. The pipeline — tumor sequencing → mutation calling → HLA binding prediction → epitope ranking → mRNA construct — is where ML has been eating the workflow for years. Most of the heavy lifting (NetMHCpan, MHCflurry, pVACseq) is already open. What's missing is a clean, reproducible end-to-end stack you can run on your own GPU cluster without pharma-grade infra.\n\nI've been stitching together a minimal viable version for a side project. Rough outline if anyone wants to replicate:\n\n**1. Somatic variant calling**\n\n```\n# tumor/normal pair → filtered VCF\ngatk Mutect2 \\\n  -R hg38.fa \\\n  -I tumor.bam \\\n  -I normal.bam \\\n  -tumor TUMOR \\\n  -normal NORMAL \\\n  -O somatic.vcf.gz\n```\n\n**2. HLA typing from RNA-seq (or WES)**\n\n```\noptitype --rna -i tumor_rna.fastq.gz -o hla_typing.tsv\n```\n\n**3. Neoantigen prediction + ranking**\n\n```\n# pVACseq wrapper — handles NetMHCpan binding, expression filter, clonal fraction\npvacseq run \\\n  somatic.vcf.gz \\\n  hla_typing.tsv \\\n  sample_id \\\n  output_dir \\\n  -e 8,9,10,11 \\\n  --allele-specific-binding-threshold 0.5 \\\n  --top-score-metric lowest\n```\n\n**4. Construct design**\n\nFeed top 20-30 epitopes into your mRNA backbone (5' UTR, CDS optimization, poly-A tail). I'm using a custom codon optimizer + UTR selector trained on ribosome profiling data — happy to share the notebook if there's interest.\n\n**5. In silico immunogenicity sanity check**\n\nRun the final construct through MHCflurry 2.0 presentation score + TCRdist similarity to known immunogenic epitopes. Filters out ~40% of candidates that look good on binding alone.\n\nThe bottleneck isn't prediction accuracy anymore — NetMHCpan-4.1 hits AUC 0.92 on presented ligands. It's **manufacturing turnaround**. Moderna quotes 6-8 weeks from biopsy to vial. Academic cores run 12-16. If you're building in this space, the lever is automating the QC/release pipeline (residual DNA, dsRNA, integrity) not the ML.\n\nAnyone running a similar stack? Curious what HLA typing method you trust — OptiType, HLA-LA, or something newer.\n\n[Next Seven Claude AI levels that actually matter for real work →](/en/threads/6959/)", "url": "https://wpnews.pro/news/ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone", "canonical_source": "https://promptcube3.com/en/threads/6982/", "published_at": "2026-08-19 22:43:02+00:00", "updated_at": "2026-08-19 23:15:20.010649+00:00", "lang": "en", "topics": ["artificial-intelligence", "machine-learning", "ai-research", "ai-tools"], "entities": ["Moderna", "NetMHCpan", "MHCflurry", "pVACseq", "OptiType", "HLA-LA", "GATK Mutect2"], "alternates": {"html": "https://wpnews.pro/news/ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone", "markdown": "https://wpnews.pro/news/ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone.md", "text": "https://wpnews.pro/news/ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone.txt", "jsonld": "https://wpnews.pro/news/ai-designed-neoantigen-vaccines-just-hit-a-phase-3-milestone.jsonld"}}