Detecting qualitative changes in biological systems

Abstract

Currently, most diseases are diagnosed only after significant disease-associated transformations have taken place. Here, we propose an approach able to identify when systemic qualitative changes in biological systems happen, thus opening the possibility for therapeutic interventions before the occurrence of symptoms. The proposed method exploits knowledge from biological networks and longitudinal data using a system impact analysis. The method is validated on eight biological phenomena, three synthetic datasets and five real datasets, for seven organisms. Most importantly, the method accurately detected the transition from the control stage (benign) to the early stage of hepatocellular carcinoma on an eight-stage disease dataset.

Machine learning classifies predictive kinematic features in a mouse model of neurodegeneration

“…However, one of the main limitations of this study is that we did not establishing a direct relationship between the load of hAPP in the neural system and the severity of kinematic feature patterns that we observed. Moreover, sequential measurement of time-series data from more age groups could increase the resolution of detection of gene-related progression of motor disorders 91 . Despite the difference between the quadrupedal stepping and bipedal stepping, the patterns of hindlimb kinematics are relatively conserved due to the involvement of hind limbs body weight support, and similar analyses may be performed in humans.…”