One platform, endless possibilities

Our ImmunoEngine for personalized pipeline is IVD-ready, involves whole genome target discovery, and is best-in-class in target prioritization. It fits personalized clinical manufacturing timelines of vaccine and T-cell therapeutics, delivering results from whole genome sequencing to neoantigen selection and product design in one week.

We offer an end-to-end workflow with a turnaround of 5 business days:

• Variant expression analysis
• Translation analysis
• Presentation prediction: identification of MHC-presented epitopes on both MHC-I and MHC-II molecules
• Immunogenicity prediction: identification of presented epitopes recognized by both CD4 and CD8 immune cells
• Final epitope selection: based on antigen score, clonality, functional relevance, and immune-escape likelihood
• Data visualization: web-based, comprehensive solution for navigating complex datasets
• myCONSTRUCT: next-generation string-of-beads design (optional)
• myRNA: next-generation codon optimization (optional)

Our ImmunoEngine pipeline is also optimized for large cohort analyses allowing shared target discovery in specific cancer patient populations. We perform custom discovery for vaccine and T-cell therapeutics developers, identifying and prioritizing MHC-bound targets using our advanced prediction models. Through our expertise in whole genome target discovery, we find more, better, and novel (dark) targets, including camyotopes, increasing the likelihood for broad population coverage.

We offer large-scale patient cohort analyses with an average turnaround time of 2–3 months:

• Genome-wide association studies → linking genetic variation to phenotypes such as treatment response, disease progression, or therapeutic targets
• Differential expression and translation analysis
• Presentation prediction: identification of MHC-presented epitopes on both MHC-I and MHC-II molecules
• Immunogenicity prediction: identification of presented epitopes recognized by both CD4 and CD8 immune cells
• Coverage optimization: matching antigen targets with population of interest for optimal coverage
• Final epitope selection: based on antigen score, clonality, functional relevance, and immune-escape likelihood
• myCONSTRUCT: next-generation string-of-beads design (optional)
• myRNA: next-generation codon optimization (optional)

Our ImmunoEngine-powered discovery pipeline myPATHOGEN is optimized for the identification of highly conserved and immunogenic epitopes across pathogen strains, combining AI-guided antigen selection, conservation analysis, immunogenicity ranking, population coverage, and translation likelihood.

We offer an end-to-end flow with a turnaround time of 8 weeks:

• Genome-wide association studies → linking pathogen genetic variation to phenotypes such as virulence, drug resistance, or immune evasion
• Conservation analysis: prioritization of immunogenic epitopes within conserved regions
• Presentation prediction: identification of surface-presented epitopes in populations infected with different strains
• Immunogenicity prediction: identification of presented epitopes recognized by immune cells
• myCONSTRUCT: next-generation string-of-beads design (optional)
• myRNA: next-generation codon optimization (optional)

mySELF offers accurate prediction of immunogenic and tolerogenic self-antigens. It is an autoimmune-focused adaptation of our best-in-class immunogenicity prediction engine, designed to quantify the immunogenic potential of self-derived peptides across both CD8 and CD4 contexts. It enables rational prioritization of pathogenic candidates and supports region-level selection for tolerance-oriented strategies by moving beyond MHC-binding and integrating presentation as well as T-cell activation features into a unified scoring framework.

We offer large-scale patient cohort analyses with an average turnaround time of 2–3 months:

• Epitope prioritization: robust CD8 epitope prioritization for pathogenic mechanism and target discovery
• CD4 region assessment: region-based CD4 analysis aligned with tolerance-oriented concepts
• Final epitope selection: rational narrowing of large autoimmune candidate spaces
• Data integration: integration with HLA, expression, immunopeptidomics, and TCR datasets

myEPITOPE delivers accurate, scalable, biologically grounded immunogenicity prediction — enabling more potent immunotherapeutics. Built around our first-in-class high-precision immunogenicity prediction tool neoIM, it estimates the likelihood that MHC-I–presented epitopes elicit CD8⁺ T-cell responses. The algorithm outperforms existing tools and de-risks preclinical validation significantly.

We offer immunogenicity screening of epitopes with a turnaround time of 1 week:

• Presentation prediction: identification of target epitopes likely to be presented at the cell surface across various patient populations
• Immunogenicity hotspot identification: identification of high-impact immunogenic epitope clusters across the target sequence
• Similarity-to-self analysis: isolation of the target epitopes least related to the canonical human proteome
• Coverage estimation: fine-grained predictions of target population coverage for a select set of immunogenic epitopes
• myCONSTRUCT: next-generation string-of-beads design (optional)
• myRNA: next-generation codon optimization (optional)

myCONSTRUCT offers rational amino acid construct design reducing unwanted junctional epitopes, minimizing unintended immunodominance, and preserving therapeutic focus. By integrating proteasomal processing, MHC presentation, immunogenicity prediction, and population-aware aggregation into a unified risk framework, myCONSTRUCT transforms string-of-beads design from heuristic assembly to predictive engineering of multi-epitope therapeutics.

We offer string-of-beads design optimization with a turnaround time of 1 week:

• Junctional peptide generation: enumeration of all peptide fragments that could arise across the junction following intracellular processing
• Processing and presentation modeling: evaluation of each junction for proteasomal likelihood, MHC class I (and optionally class II) presentation probability, and allele-specific binding and presentation features
• Immunogenicity scoring: estimation of the probability a presented peptide will elicit a T-cell response
• Integrated junction risk aggregation: risk score aggregation across alleles, weighted by allele frequency, and summarized at construct level
• Design optimization: order refinement, spacer selection, and construct-level junction penalty minimization

myRNA enables proactive, integrated construct engineering in a single unified framework. It is a next-generation codon optimization engine that integrates structural stability and codon usage with comprehensive, constraint-aware design. Pair with myCONSTRUCT for end-to-end construct optimization. It enables scalable, multi-epitope RNA therapeutics codon optimization while preserving fixed regions and incorporating biologically relevant sequence safeguards.

We offer codon optimization with a turnaround time of 1 week:

• Multi-epitope construct design support: coding and non-coding regulatory regions are treated as a single structurally interdependent molecule
• Practical therapeutic design constraints integration: including stem size control, restriction site avoidance, and degradation hotspot mitigation
• Modified nucleotide optimization in-silico: evaluation of stability effects for pseudouridine and 5-methylcytosine
• Runtime–accuracy flexibility: rapid exploration or near-exact optimization depending on user needs, via adjustable beam width

We analyze large-scale immuno-genomic datasets to extract actionable biological insights including biomarker identification, response prediction, trial design optimization, data imputation, and cross-dataset interpretation to support data-driven decision-making. View our partnered programs leveraging these insights.

We offer support with:

• Deep investigation of patient data: omics, TCRseq, ELISpot, ICS, MSD 30-plex, IHC, immunophenotyping, Nanostring
• ctDNA patient monitoring and follow-up
• Tumor-specific mutational profiling
• Comparison vs digital genomic twin with similar immuno-genomic profile
• Biomarker identification based on clinical trial results
• Pre-treatment biomarkers for patient stratification
• Early biomarkers correlating with clinical response
• Patient-centric translational data analysis
• Therapy response prediction: TMB, MSI, immunogenic neoantigen load
• Tumor heterogeneity, TME classification
• AI-driven multi-parametric analysis
• Experimental design optimization