The development, evaluation and use of freshly isolated renal proxinal tubule systems in the fischer rat

The investigation of renal pathophysiology and toxicology has traditionally been advanced by the development of increasingly defined and refined in vitro preparations. This study has sought to develop and evaluate various methods of producing pure samples of renal proximal tubules (PTs) from the Fischer rat. The introduction summarised the most common in vitro preparations together with the parameters used to monitor viability - particularly with regard to toxic events. The most prevalent isolation methods have involved the use of collagenase to produce dissociation of the cortex. However, the present study has shown that even the mildest collagenase treatment caused significant structural damage which resulted in a longevity of only 3hr in suspension. An alternative mechanical isolation technique has been developed in this study that consists of perfusion loading the renal glomeruli with Fe304 followed by disruption of the cortex by homogenisation and sequential sieving. The glomeruli are removed magnetically and the PTs then harvested by a 64μM sieve. PTs isolated in this way showed a vastly superior structural preservation over their collagenase isolated counterparts; also oxygen consumption and enzyme leakage measurements showed a longevity in excess of 6hr when incubated in a very basic medium. Attempts were then made to measure the cytosolic calcium levels in both mechanical and collagenase isolated PTs using the fluorescent calcium indicator Fura. However results were inconclusive due to significant binding of the Fura to the external PT surfaces. In conclusion, PTs prepared by the present mechanical isolation technique exhibit superior preservation and longevity compared with even the mildest collagenase isolation technique and hence appear to offer potential advantages over collagenase isolation as an in vitro renal system.

Practical methods of tracking of nonstationary time series applied to real-world data

In this paper, we discuss some practical implications for implementing adaptable network algorithms applied to non-stationary time series problems. Two real world data sets, containing electricity load demands and foreign exchange market prices, are used to test several different methods, ranging from linear models with fixed parameters, to non-linear models which adapt both parameters and model order on-line. Training with the extended Kalman filter, we demonstrate that the dynamic model-order increment procedure of the resource allocating RBF network (RAN) is highly sensitive to the parameters of the novelty criterion. We investigate the use of system noise for increasing the plasticity of the Kalman filter training algorithm, and discuss the consequences for on-line model order selection. The results of our experiments show that there are advantages to be gained in tracking real world non-stationary data through the use of more complex adaptive models.