Search papers, labs, and topics across Lattice.
This paper introduces a novel framework for estimating the Kullback-Leibler (KL) divergence in function space, addressing the challenges of trajectory inference from snapshot data where only independent samples are available. The proposed estimator is validated against a benchmark suite, demonstrating its accuracy in approximating the analytic KL divergence, and is applied to both synthetic and real single-cell RNA sequencing datasets. The findings reveal that traditional evaluation metrics can be misleading, while the functional KL provides a more coherent and principled approach for assessing trajectory inference methods, particularly in scenarios with sparse data.
Current evaluation metrics for trajectory inference can mislead researchers, but functional KL divergence offers a clearer, more reliable comparison of methods in sparse data conditions.
Trajectory Inference (TI) seeks to recover latent dynamical processes from snapshot data, where only independent samples from time-indexed marginals are observed. In applications such as single-cell genomics, destructive measurements make path-space laws non-identifiable from finitely many marginals, leaving held-out marginal prediction as the dominant but limited evaluation protocol. We introduce a general framework for estimating the Kullback-Leibler divergence (KL) divergence between probability measures on function space, yielding a tractable, data-driven estimator that is scalable to realistic snapshot datasets. We validate the accuracy of our estimator on a benchmark suite, where the estimated functional KL closely matches the analytic KL. Applying this framework to synthetic and real scRNA-seq datasets, we show that current evaluation metrics often give inconsistent assessments, whereas path-space KL enables a coherent comparison of trajectory inference methods and exposes discrepancies in inferred dynamics, especially in regions with sparse or missing data. These results support functional KL as a principled criterion for evaluating trajectory inference under partial observability.
