摘要Recent technological advances mean that samples from animal experiments may be analysed more cheaply,more easily and with a much greater return of data than previously.Research groups are frequently faced with a choice of continuing to use established technology in which they may have made a significant investment of time and resources,and have significant amounts of reference data,or switching to new technology where reference data may be limited.Apart from cost,the choice needs to be based on a comparison between the increase in data available from future experiments by switching and the value of comparison with reference data from historical experiments analysed with earlier technology.One approach to this problem is to ensure that sufficient quantity and variety of samples are taken from each experiment and appropriately stored to allow re-establishment of a sufficiently large reference set and to avoid the need to repeat animal experiments.The establishment of'biobanks'of experimental material will require funding for infrastructure,consistent storage of metadata and,importantly,horizon-scanning to ensure that samples are taken appropriately for techniques which will become accessible in future.Such biobanks are a recognised resource in human medicine,where the value of samples increases as more analysis is carried out and added to the metadata.