Autoimmune Diseases — Antigen-Specific mRNA Immune Tolerance
Research ownership & attribution: Rami M. Alkhaleeli — Alkhaleeli BioAI LLC
Scientific Concept
Autoimmune diseases arise from lost tolerance to self-antigens, and current treatment relies mainly on broad immunosuppression. This manuscript proposes an antigen-specific tolerance platform: AI-guided autoantigen selection and sequence optimization, in silico biosafety screening, recombinant-protein reference production and mRNA–lipid nanoparticle formulation, aimed at expanding regulatory T cells and inducing effector T-cell anergy without global immunosuppression. Cardiac myosin heavy chain α (MYH6) is used as a model antigen for autoimmune myocarditis; other antigens (COL2A1, MBP, GAD65 and others) are proposed through antigen substitution. The author states that this is a research proposal requiring extensive experimental validation, and that the mechanistic rationale of the in silico CRISPR screen for mRNA needs clarification.
Scientific Diagram
Scientific diagram not provided for this case. The documented analysis is presented in the sections below.
Design Process
- 1
Autoantigen identification — disease-relevant, tissue-restricted antigens (e.g., MYH6 for autoimmune myocarditis) are selected from genomic and protein databases.
- 2
AI-guided sequence optimization — codon usage, GC content, CpG reduction and secondary structure are optimized for the model antigen (MYH6 epitope region, aa 400–800).
- 3
Computational biosafety screening — an in silico CRISPR off-target / PAM-site screen is applied (the manuscript notes its mechanistic rationale for mRNA needs clarification).
- 4
Recombinant protein reference — a bacterial-expression workflow is described to produce a reference antigen for validation.
- 5
mRNA–LNP formulation concept — N1-methylpseudouridine mRNA in lipid nanoparticles; a dose-escalation regimen is proposed only as a conceptual starting point.
Proposed Biological Mechanism
- 1
Antigen is taken up by tissue-resident and lymph-node dendritic cells.
- 2
MHC class II presentation to CD4+ T cells is proposed to occur without danger signals.
- 3
T-cell receptor engagement without adequate co-stimulation is proposed to lead to anergy or regulatory differentiation.
- 4
Antigen-specific FoxP3+ regulatory T cells are proposed to expand and suppress effector responses.
- 5
Tolerance memory is proposed to follow — a hypothesis only, not an experimentally demonstrated effect.
All steps describe intended, modeled or proposed behavior — no effect above is experimentally proven or clinically demonstrated.
Mathematical Analysis
GC content (MYH6 design)
48.2% → 57.3%
Wild-type versus AI-optimized MYH6 epitope-region design; target range 50–60%.
CpG dinucleotides
67 → 15 (78% reduction)
Fewer CpG motifs are intended to reduce predicted TLR9 activation.
Codon adaptation index (CAI)
0.67 → 0.89
Predicted improvement in codon usage; the manuscript places it in the 94th percentile of predicted expression.
Designed construct
1,547 nt total · 1,200 nt CDS
Total designed mRNA length and coding sequence (400 amino acids, ~44 kDa encoded protein).
In silico PAM-site screen
147 sites · risk scores 2–4
Computational screen with no sites at risk score ≥7 on a 10-point scale. The manuscript notes its relevance to mRNA needs clarification, and it is not a safety demonstration.
Proposed dose escalation
1 → 10 → 50 µg, weekly × 6
Conceptual dosing starting point only; the manuscript says it requires empirical dose-finding.
Predicted HLA population coverage is a computational prediction of epitope presentation — it is not a measure of actual vaccine protection. All values originate from the supplied manuscript materials.
FASTA Sequence Viewer
FASTA — 2 sequences
Sequences exactly as printed in the manuscript. The mRNA is shown only as the first ~500 nucleotides printed there; the full construct is not shown. Theoretical designs for scientific discussion.
> sp|P13533|MYH6_HUMAN Myosin-6 (Epitope 400-800)
Length: 386 characters
> MYH6_epitope_400-800_AI_optimized_mRNA (first 500 nt as printed)
Length: 494 characters
Original Scientific Manuscript
Read-only viewing
The original manuscript PDF is available below for read-only viewing only. Downloading, copying or redistribution is not permitted without written permission from Alkhaleeli BioAI LLC.
© Alkhaleeli BioAI LLC. All rights reserved. This manuscript is the intellectual property of Alkhaleeli BioAI LLC and is provided for read-only viewing only. Downloading, copying, printing, redistribution, or any other use — in whole or in part — requires prior written permission from Alkhaleeli BioAI LLC.
Scientific Status
Autoimmune Diseases — Conceptual
These projects are scientific research designs and computational proof-of-concept studies. They are not approved medical products, experimentally validated cures, or guarantees of therapeutic effectiveness.