Vertex's 3,456-well Nanowell plates enable ultra-high throughput. Credit: VERTEX PHARMACEUTICALS

Miniaturization has been one of the triumphs in screening technology over the past decade or so, mainly because of advances in the automation of liquid handling, control software and detection systems. A few examples of ultra-high-throughput screening already exist, with Vertex in San Diego, California, for instance, performing the majority of its assays using 3,456-well microplates, in which each assay is done in a volume of just 1 µl.

Affymax, a drug-discovery firm based in Palo Alto, California, is reported to be working on a 20,000-well plate in which each well would have a volume of just 25 nl. And although most of the endeavour has been directed towards miniaturizing microplates, other ultra-high-throughput formats are being developed, the screening of microbead-attached combinatorial libraries, for instance, by companies such as Luminex in Austin, Texas, and Illumina in San Diego, California.

Although these examples are undoubtedly a taste of the future, miniaturization is not for everyone, and does not suit every purpose. Except when compound or protein amounts are critical factors, miniaturization is unlikely to make a big difference to the efficiency of primary screening, although doing assays in less time makes it easier to keep conditions standardized.

For Vertex's cell-based assays, the transition from 384- to 3,456-well plates meant that the number of cells needed dropped from 10,000–40,000 per well to about 200. This means that the company can work with cell types that are relatively difficult to get hold of, such as disease cells, says Paul Negulescu (above, inset), vice-president of discovery biology at Vertex. In principle, 3,456-well plates could be used to study single cells, but at that level the cells' 'individuality' starts to become a problem, giving uneven responses and so degrading the data.