Pub-AI: AI in Science

Driving · 2026-09-30

P2P: Cross-View Population Denoising for Unpaired Single-Cell Perturbation Response Prediction

Haojie Yang, Ran Su

arXiv:2609.34391PDF

Auto-summarized: this summary was generated by a language model from the paper’s text and has not been reviewed by an editor. Check the paper before relying on it. Benchmark numbers appear only after a human has verified them.

TL;DR

P2P predicts unpaired single-cell perturbation response at the population level by learning shared condition effects from two stochastic “views” of the same control/perturbed condition. It encodes control sets with a permutation-invariant set encoder, perturbations with a structured encoder, and blends an empirical memory with a neural residual to output population mean and gene-wise variance. Trained with cross-view supervision (no cell correspondences).

Why it matters

Perturb-seq does not provide paired control/perturbed cells, and sampling variation can confound cell-level regression. P2P uses stochastic cell-set views so that the model learns a reproducible population effect via cross-view agreement, aiming to improve effect recovery and differential gene ranking for perturbation response prediction.

Method

  • Use cross-view supervision: sample two stochastic views from each control and perturbed condition; predict the other perturbed view mean from one control view plus perturbation metadata, without requiring cell correspondences.
  • Encode control populations with a permutation-invariant set encoder (attention-weighted summary plus dispersion summary) and encode perturbation tokens with structured encoders for tokens, context, dose, and pair interactions.
  • Predict population mean with a memory-gated effect head (empirical condition-effect memory blended with a neural residual) and predict gene-wise log variance with a heteroscedastic head; optimize a cross-view objective including effect and consistency terms.
Abstract (from arXiv)

AIVC (AI Virtual Cell) is a learned simulator of cellular behavior across conditions. Predicting how a cell population responds transcriptionally to a genetic perturbation is a core task. Perturb-seq records that response by destructive sequencing, so a control cell and a perturbed cell are never observed as a pair, and cells under one condition remain heterogeneous and noisy. Regression on individual cells absorbs sampling variation into the estimated effect, whereas interpretation requires the reproducible population effect. P2P (Perturbation-to-Perturbation) takes a stochastic cell-set view as its supervision unit. Two views drawn from the same condition share a reproducible population effect and differ by view-specific variation. A permutation-invariant set encoder summarizes the control population, a structured encoder represents perturbation tokens, cellular context, dose, and combination interactions, and a gate blends empirical condition-effect memory with a neural residual. A heteroscedastic head predicts the population mean and gene-wise response variance. Under one protocol and five seeds, P2P attains the lowest expression RMSE and the highest Effect Pearson, DEG F1, and DEG average precision on each of Adamson, Norman, Replogle K562, and Replogle RPE1 relative to GenePert, LinearPert, SLIM, Scouter, and scPILOT. On Replogle K562, Effect Pearson rises from 0.643 to 0.702 and DEG F1 rises from 0.067 to 0.178 relative to Scouter, the strongest baseline on both metrics.

Related papers