Bisphenol A (BPA), a typical endocrine disruptor and a contaminant of emerging concern (CECs), has detrimental im-pacts not only on the environment and ecosystems, but also on human health. Therefore, it is essential to investigate the degrading processes of BPA in order to diminish its persistent effects on ecological environmental safety. With this objective, the present study reports on the effectiveness of biotic/abiotic factors in optimizing BPA removal and eval-uates the kinetic models of the biodegradation processes. The results showed that BPA affected chlorophyll a, super-oxide dismutase (SOD) and peroxidase (POD) activities, malondialdehyde (MDA) content, and photosystem intrinsic PSII efficiency (Fv/Fm) in the microalga Chlorella pyrenoidosa, which degraded 43.0 % of BPA (8.0 mg L-1) under gen-eral experimental conditions. The bacteria consortium AEF21 could remove 55.4 % of BPA (20 mg L-1) under orthog-onal test optimization (temperature was 32 degrees C, pH was 8.0, inoculum was 6.0 %) and the prediction of artificial neural network (ANN) of machine learning (R2 equal to 0.99 in training, test, and validation phase). The microalgae-bacteria consortia have a high removal rate of 57.5 % of BPA (20.0 mg L-1). The kinetic study revealed that the removal pro-cesses of BPA by microalgae, bacteria, and microalgae-bacteria consortia all followed the Monod's kinetic model. This work provided a new perspective to apply artificial intelligence to predict the degradation of BPA and to understand the kinetic processes of BPA biodegradation by integrated biological approaches, as well as a novel research strategy to achieve environmental CECs elimination for long-term ecosystem health.
