To explore the feasibility of creating a bank of stem cells, Yamanaka's team tested the tissue type of 107 Japanese volunteers. People vary highly in the genes that code for cell surface molecules called human leukocyte antigens (HLA), and the closer these antigens match between donor and recipient, the less likely the recipient's immune system is to reject the transplanted tissue. The researchers discovered that two of the volunteers were homozygous--meaning they carried two matching copies--of each of three HLA genes. Because of this rare genotype, these people could serve as donors to anyone with those three antigens--nearly one-third of the volunteer group. According to a collaborator's calculations, just 50 different HLA types could cover 90 percent of the Japanese population. The researchers have since made iPS cells from these individuals and shown that they behave normally and can be differentiated into numerous cell types.

Roger Pederson, a stem cell biologist at Cambridge University, in the U.K., warns that even this level of matching would require some immunosuppression because of slight HLA mismatches. "But it's preferable to making lines for everyone who needs one," says Pederson, who proposed the idea of a stem cell bank in 2005.

It's not yet clear how difficult it would be to create a similar tissue-matched bank for more diverse populations. "The Japanese are a particularly homogeneous population, and I wonder if it would be possible to do a similar screen in the U.S.," says Loring. Because Americans are ethnically heterogeneous, there are likely fewer homozygotes, she says.

In 2005, Pederson's team analyzed HLA types for cadaver organ donors and for people on the organ donation waiting list in the U.K., calculating that 150 lines of stem cells would provide a beneficial match for about 85 percent of those on the list. Ten donors homozygous for common HLA types would provide a beneficial match for about 67 percent. The researchers aren't yet creating such a bank, but they are analyzing the best way to create a clinically useful bank of cells.

Population-based stem cell banking is still in its early days. (Banks of newborns' umbilical cord blood, often stored in private facilities by parents, represent a different kind of banking. The stem cells from this blood are useful for a limited set of rare diseases, and they cannot be expanded or differentiated the way iPS cells can.) While scientists are studying ways to use iPS cells as disease treatments, none have yet been tested in clinical trials. In addition, these cells aren't yet suitable for human transplant; the way they are currently made may carry some risk of cancer. "We still haven't determined which type of induction is best, but after solving the technical issues, creating this kind of bank would be feasible," says Yamanaka.