MOFA {SMRUCC.genomics.Analysis.Microarray.MultiOmics.MOFA}

.NET clr documentation

MOFA

Description

Multi-Omics Factor Analysis-a framework for unsupervised integration of multi-omics data sets

Declare

            
# namespace SMRUCC.genomics.Analysis.Microarray.MultiOmics.MOFA
export class MOFA {
   # Active factor indices (after pruning)
   ActiveFactors: boolean;
   # ARD precision α^m_k (M × K). Large α → factor k is suppressed in view m. Updated during inference; factors with large α across all views are pruned.
   Alpha: Double[,];
   # Whether training has converged
   Converged: boolean;
   # ELBO history (one entry per iteration)
   ElboHistory: iterates(Double);
   GlobalSampleIds: string;
   # Current number of active factors K (may decrease due to pruning)
   K: integer;
   # Number of views M
   M: integer;
   # Number of global samples (union of all views' samples)
   N: integer;
   Options: MOFAOptions;
   # Per-feature spike-and-slab mixing probability π^m_d ∈ [0,1] (D_m × 1 per view). π close to 1 → feature d is "active" (drawn from slab); π close to 0 → feature d is "inactive" (drawn from spike).
   Pi: iterates(Double[]);
   # Per-view, per-feature noise precision τ^m_d (D_m × 1 for each view). Gaussian noise model: ε ~ N(0, 1/τ).
   Tau: iterates(Double[]);
   Views: iterates(DataView);
   # Weight matrices W^m (one per view), each (D_m × K)
   W: iterates(Tensor);
   # Factor matrix Z (N × K) — shared across all views
   Z: Tensor;
}

        

.NET clr type reference tree

1. use by property member Alpha: Double
2. use by property member ElboHistory: iterates(Double)
3. use by property member Options: MOFAOptions
4. use by property member Pi: iterates(Double)
5. use by property member Tau: iterates(Double)
6. use by property member Views: iterates(DataView)
7. use by property member W: iterates(Tensor)
8. use by property member Z: Tensor

---