Developing physiologically compatible electron donors for reductive dechlorination by Shewanella oneidensis MR-1 using machine learningDeveloping physiologically compatible electron donors for reductive dechlorination by Shewanella oneidensis MR-1 us

The targeted biological stimulation of carbon sources presents considerable potential for enhancing the dehalogenation efficiency at sites contaminated with halogenated hydrocarbons. However, the release of organic carbon in the natural environment is an extremely complex process, involving complex interactions such as carbon source types, electron transfer, and microbial metabolism. Traditional methods are difficult to effectively develop the precise carbon source materials to accelerate microbial reductive dehalogenation. In this study, machine learning (ML) approaches were employed, using the composition of carbon sources, water quality conditions and biostimulatory genomic data as input variables, and dechlorination effect as output to elucidate the biocompatibility between carbon source materials and functional strain Shewanella oneidensis MR-1.