Depending on how successful interpretation of the personal genome becomes, any method that makes the process easier for collecting the DNA is progress. That’s one way of saying that the future of personal genome medicine depends on the research that finds the links between genes and disease, and how easy and inexpensive it is to get a personal genome reading. One obvious trend is that the equipment for DNA sequencing continues to become less expensive and more powerful, something like the progression followed by computers over the last decades. Research continually finds links between specific genes and human health problems, although the validity of the interpretations (cause and effect) is in dispute. Now a new collection technique promises to make analysis of fetal DNA easier and less problematic.

The approach, developed by Jacob Kitzman and Matthew Snyder at the University of Washington (Seattle, USA), uses a sample of maternal blood taken at about 18 weeks into a pregnancy and a sample of paternal saliva to do genome sequencing and then run the combined genomes (mother, father, fetus) to make a composite and highly accurate reading of the fetus’ genome.

As presented in the journal Science: Translational Medicine [06 June 2012, paywalled, Noninvasive Whole-Genome Sequencing of a Human Fetus] this new approach avoids the difficulty and potential dangers of extracting a fetal DNA sample from inside the mother’s womb. The non-invasive technique relies on the fact that the mother’s blood contains a certain percentage of fetal DNA, which can be isolated and extracted. Several other research groups are also working with this source of fetal DNA, but the Kitzman/Snyder approach also uses the mother’s and father’s genomes to increase the accuracy of assessing the baby’s genome. The claim is 98% percent accuracy.

One of the difficulties in assessing the fetal genome is that in addition to inheriting genes from the mother and father, there are almost certain to be a small number of mutations unique to the fetus. Since these mutations account for a high percentage of genetic disorders, their detection is crucial. It is also, naturally, the hardest part of obtaining accuracy in a genetic sequence of a fetus. Kitzman and Snyder have developed complex software routines, some of them based on statistics and economics, to analyze the sequences looking for mutations. At the moment they are able to discover about 39 out of 44 de novo (not found in the mother or father) mutations. Not bad, but neither scientist considers it good enough.

This difficult identifying and interpreting genetic information, which includes ‘normal’ genes as well as mutations, remains the most formidable barrier for the eventual acceptance of using the personal genome for genetic and medical analysis.