Search papers, labs, and topics across Lattice.
This paper introduces PerturbMap, a novel method for predicting missing recipient-context effects in single-cell perturbation atlases by integrating a recipient-local low-rank base with source-to-recipient ridge experts. The approach addresses the challenge of transferring query-specific experimental evidence without losing accuracy, which is crucial for effective cellular context analysis. On the Perturb-CITE-seq melanoma cohort, PerturbMap demonstrates a 4.1% improvement in full-effect mean squared error (MSE) compared to the baseline, outperforming several existing methods and closely matching a centralized reference model.
PerturbMap achieves a 4.1% reduction in prediction error for single-cell perturbation responses, significantly enhancing cross-context transfer accuracy.
Single-cell perturbation atlases rarely measure every intervention in every cellular context: a query perturbation is often observed in one or more source contexts but missing in the recipient context where its effect is needed. Ignoring those measured responses discards query-specific experimental evidence, whereas copying or weakly calibrating them across contexts risks transferring the wrong signal. We propose PerturbMap, which predicts a missing recipient-context effect by combining a recipient-local low-rank base with accepted proposals that transport the same perturbation's measured source responses through source-to-recipient ridge experts fit on paired training perturbations, with proposal weights determined by route reliability estimated on validation anchors. On the Perturb-CITE-seq melanoma cohort, PerturbMap improves full-effect MSE by 4.1\% over a recipient-local low-rank base and achieves lower MSE than FedAvg, zero-response, raw-copy, calibrated-copy, and identity-shuffled affine controls. It remains within $2.82\times10^{-6}$ MSE of our centralized token-matched pooled reference, which uses a stronger training interface. A condition-mean specificity diagnostic shows the same direction: same-recipient top-10 counterpart retrieval by cosine increases from 74.5\% for the low-rank base to 80.5\% for PerturbMap.
