This talk will present some data analysis and machine learning methods used in dark matter and neutrinoless double beta decay experiments that could be applied to signal processing of LUNA, enhancing the sensibility in some nuclear reaction channels. It will also be discussed the results of nucleosynthesis simulations using Skynet and nuclear reactions using Fluka. Finally, it will be presented a possible neutron flux monitoring using silicon detectors.