ADVANCED PIPELINES

Multi-Step Workflows for Deeper Analysis

These pipelines chain multiple models together to tackle bigger questions — useful when a single structure prediction isn't enough on its own.

Neoantigen & Receptor Design Pipeline

Explore how patient-specific mutations might create new targets for immune therapies, and how a designed receptor would bind to them.

Pipeline steps
  1. 1.Gamma models the receptor-target interaction
  2. 2.Delta evaluates how a candidate molecule would bind to the assembly
  3. 3.Results are scored and ranked automatically
Variant Triage Pipeline

Screen a large list of gene variants quickly to identify which ones are most likely to be structurally significant, then get detailed structural analysis on the top candidates.

Pipeline steps
  1. 1.Alpha screens all variants for a fast first-pass score
  2. 2.Top-ranked variants are automatically queued for detailed analysis with Delta/Beta
  3. 3.A summary table ranks all variants by predicted impact
Resistance & Inhibitor Design Pipeline

Investigate how a resistant version of a target protein differs from the normal version, and test how well candidate molecules bind to each. Candidate molecules are drawn from real public compound databases (ChEMBL); binding simulation against your target requires a connected computation backend (see platform roadmap).

Pipeline steps
  1. 1.Beta docks candidate molecules against the normal target
  2. 2.The same molecules are docked against the resistant variant
  3. 3.Side-by-side comparison shows which candidates retain binding despite the resistance mutation
Vaccine Design Pipeline

Model a viral surface protein in different conformations and identify which regions are the best targets for a vaccine to focus on.

Pipeline steps
  1. 1.Delta + Beta model the viral protein's different shapes
  2. 2.Gamma identifies and ranks candidate antibody-recognition regions
  3. 3.Results highlight the most promising target regions
Protein Misfolding Pipeline

Explore the range of shapes a misfolding-prone protein can take, and see where potential treatments might intervene.

Pipeline steps
  1. 1.Delta + Beta model multiple possible conformations of the protein
  2. 2.Results are compared to known experimental data where available
  3. 3.Candidate intervention points are flagged on the structure
Cardiac Structure Pipeline

Model ion channels and lipid-transport proteins relevant to heart conditions, and see how specific mutations affect their structure.

Pipeline steps
  1. 1.Delta models the full protein assembly
  2. 2.Alpha models the channel structure with the mutation highlighted
  3. 3.A summary explains the likely functional impact
Note: very large proteins are modeled in sections rather than all at once for best results.
Pediatric Variant Pipeline

Quickly screen a child patient's variant list to flag which ones are most likely to be clinically relevant, with structural explanations suitable for sharing with families.

Pipeline steps
  1. 1.Alpha rapidly screens all variants from the input list
  2. 2.Each variant gets a likelihood classification (likely pathogenic / likely benign / uncertain)
  3. 3.A plain-language summary is generated for each flagged variant
Longevity & Aging Pathways Pipeline
Exploratory

Explore proteins involved in aging-related pathways (DNA repair, cellular senescence, mitochondrial function) and screen candidate molecules against them.

Pipeline steps
  1. 1.Delta + Beta model key aging-pathway proteins
  2. 2.Beta + Gamma dock candidate molecules against these targets
  3. 3.Alpha screens related variants from reference datasets for structurally interesting patterns