Single Cell Data Analysis Algorithms

National Institute Of Diabetes And Digestive And Kidney Diseases

Investigators

Abstract

With advances in single-cell techniques, collecting a large quantity of data has become more accessible and efficient. In contrast, the increased complexity of data has made it more challenging to unravel underlying biological mechanisms. Thus, it is critical to develop novel computational methods capable of dealing with such complexity and of providing some predictive deductions from the data. Fowlkes et al. (Cell, 2008) published a set of gene expressions measured from 6078 Drosophila blastoderm during six different time cohorts. Out of 95 genes and four proteins, only 27 of them had complete temporal information from all the cells, while the rest were measured only in a subset of cells. To impute the missing data, we trained and tested neural networks with one hidden layer on the complete 27 genes as predictors and the genes that were measured only in subsets of cells as targets. With the trained neural network, we imputed the missing gene expressions. To test the imputation methods performance, we arbitrarily selected three genes from the complete 27 genes and randomly removed time points from their gene profiles. Then, the missing values were imputed using the same method. The medians of the imputed values were compared to those of the observed values and showed negligible differences. In this study, we present a novel method for the inference of a gene network using a new approach to least absolute deviation regression after using neural networks for imputing missing data. Our iterative regression algorithm for inferring a mechanistic gene network from single-cell data is especially suited to overcoming problems posed by measurement outliers. Using this regression, we infer a developmental model for the gene dynamics in Drosophila melanogaster blastoderm embryo. Our results show that the predictive power of the inferred model is higher than that of other models inferred with least squares and ridge regressions. As a baseline for how well a mechanistic model should be expected to perform, we find that model predictions of the gene dynamics are more accurate than predictions made with neural networks of varying architectures and complexity. This holds true even in the limit of small sample sizes. We compare predictions for various gene knockouts with published experimental results, finding substantial qualitative agreement. We also make predictions for gene dynamics under various gene network perturbations, impossible in non-mechanistic models.