Method Choice, Not Biology, Determines In Silico Perturbation Results: A Systematic Evaluation of Eight Methods Across Four Datasets

Most in silico perturbation methods for single-cell transcriptomics have been validated only on individual datasets, leaving their reliability and generalizability unknown. Through systematic cross-method, cross-dataset benchmarking of eight methods spanning six mathematical frameworks across four datasets, we find that six of eight methods--including widely used VAE-based and tensor decomposition approaches--fail to produce detectable transcription factor (TF)-to-pathway signals. Only CellOracle and DDIM consistently detected TF-to-glycolysis directional regulation. Cross-pathway analysis in PBMC monocytes revealed biologically coherent TF-pathway associations beyond glycolysis (SPI1[->]glycolysis 4.4x enrichment, FOS[->]AP-1 targets 4.4x), with SOX9 serving as a biological specificity control (no pathway enrichment). Method choice alone could reverse biological conclusions: DDIM and scTenifoldKnk rankings were significantly anti-correlated ({rho}=-0.811, p=0.027). CRISPRi Perturb-seq validation in K562 cells confirmed TF knockdown suppresses glycolysis gene expression (JUN {delta}=-1.72, CEBPB {delta}=-1.59, SPI1 {delta}=-1.57, FOS {delta}=-0.70), but CellOracle-predicted perturbation directions did not match experimental directions (40.9% agreement, not different from chance), revealing a fundamental gap between steady-state correlation and causal perturbation. Diagnostic analyses using VAE latent space profiling, correlation distribution comparison, and gene-gene graph analysis identified distinct failure modes in unsuccessful methods: VAE latent space competition (STAT3 signal-to-noise 0.44 vs. SPI1 4.25), correlation noise (TF-glycolysis |r|=0.038 indistinguishable from background |r|=0.047), and graph non-specificity (0.84x enrichment). A controlled ablation experiment showed that adding a GRN prior to DDIM did not improve target recall (delta=0 for all TFs), confirming that performance differences are multi-factorial. These findings establish preliminary guidance for method selection, including cross-pathway validation, direction-aware benchmarking, and minimum data requirements ([≥]500 cells, [≥]1,000 HVGs).