A Performance Evaluation Model of Single-Cell Pseudotime Trajectory Inference Algorithms

The study on single-cell pseudotime trajectory is of great significance to the exploration of the environmental factors of life and diseases. The large scale and complexity of single-cell data make the single-cell pseudotime trajectory algorithms face great challenges. A performance evaluation model is proposed to measure the performance of existing pseudotime trajectory inference algorithms and mine the problems existing in the inference algorithms in order to promote the improvement of the inference algorithms. Under the condition of given original single-cell data, the model uses the Spearman correlation coefficient to evaluate the performance of the inference algorithms from noise resistance and robustness. Experiments on the algorithms Monocle2 and Scout were conducted to analyze the application effect of the model.