Bioproducts assessment using quantitative image analysis in a mixed microbial culture fed with used cooking oil

The use of image analysis for the quantification of polyhydroxyalkanoates (PHAs) and triacylglycerides (TAGs) produced by a mixed microbial culture fed with used cooking oil was studied. Three different stains were used, the Sudan Black B (SBB), the Nile Blue A (NBA) and the BODIPY™ 493/503. Acquired images were analysed using Matlab R2024b software, and bioproduct (PHA and TAG) accumulation was predicted through partial least squares (PLS) regression. For PHA prediction, the three stains showed good performance, with NBA achieving the best results with a R2 of 0.93 and a residual prediction deviation (RPD) of 3.57. However, for the prediction of TAG accumulation all stain resulted unreliable, though SBB showed the best results with a R2 of 0.68 and a RPD of 1.44. These findings demonstrate the feasibility of using different stains for quantification of the accumulated PHA inside microbial cells, but further research is needed to enhance TAG prediction accuracy.