Our paper titled "Global sensitivity analysis with limited data via sparsity-promoting D-MORPH regression: Application to char combustion" has been accepted for the publication in Journal of Computational Physics. The co-authors are Elle Lavichant, a master's student at UC San Diego, and Prof. Boris Kramer, who supervised my postdoctoral position at UC San Diego. In this study, we present a novel D-MORPH regression to train a surrogate that minimizes the required training data. This method first computes a sparse Lasso solution and uses it to define the cost function. A subsequent D-MORPH regression minimizes the difference between the D-MORPH and Lasso solution, resulting in a surrogate more robust to input variations and accurate with limited training data. This efficiency is demonstrated using a char combustion model. The new method requires only 15% of the training data compared to conventional regression.