By Joel Warner
By Michael Roberts
By Alan Prendergast
By Michael Roberts
By Michael Roberts
By Amber Taufen
By Patricia Calhoun
By William Breathes
Ever since the center began doing paternity testing, in 1985, business has increased steadily. In 1990, the Analytical Genetic Testing Center performed ninety tests. Last year it did 1,020. The biggest increase came in 1993, when the center won the business of the City and County of Denver's Department of Social Services. Because every effort is made to find a father in the thousands of child support cases it handles each year, Denver's business alone would keep the testing center hopping.
Such government work makes up the bulk of the center's non-forensic business. However, parents and children bring in blood samples for all sorts of reasons. Some parents want to know if they've passed on a medical condition to their kids. "One father brought in his little baby for testing because he said she was so cute he couldn't actually believe she was his," Schanfield says. "He was happy to learn she was." The oldest child the center has tested was a 57-year-old man. His father had died recently, and he wanted to know if the man was really his parent.
In the past ten years, the exclusion percentage -- the number of tests showing that the presumed father actually was not -- has stayed about the same, somewhere between 27 and 33 percent. This parallels national statistics: Affiliated members of the American Association of Blood Banks performed 280,000 paternity tests in 1999, the last year for which figures are available; in about 28 percent of those, the man being tested was shown not to be the father. While the sample population is skewed -- most parents who ask for testing have at least a private suspicion -- it is nonetheless an astounding number.
The technology used to read parents' and children's shared genetic roadmaps has improved tremendously in the past decade. In 1985, a paternity test could guarantee a 90 percent rate of probability. Today, most tests are somewhere around 99.9 percent conclusive -- a margin of error so slim as to be practically nonexistent.
Blood is the preferable agent for testing, although a cheek swab will work in a pinch. (Hair strands don't work because the cells in them are passed down only from mother to child, thus leaving no evidence of paternity.) First the blood is separated into its basic components of plasma, red blood cells and white blood cells. The pure DNA is extracted from the white cells when they are placed in a reagent that dissolves the coating around the genetic code -- like removing the candy outside to get to the inside of a Tootsie Pop.
Next, the DNA is placed in a machine that essentially photocopies its unique patterns. The multiple copies of the genetic code allow the testers to focus on the exact locations they need to study in order to determine a child's and parents' common biological landmarks.
Each strand of genetic coding carries more than 30,000 genes, far too many to sort out which came from a particular parent. Also, many genes are common to everyone -- for example, the gene that determines how a person's heart develops. So testers hoping to prove a family connection focus on about a dozen known as "nonsense genes," those that have no visible genetic expression in a person -- blond hair, say, or straight teeth -- but that nevertheless carry biological clues to a person's individuality.
From there the process is relatively low-tech, despite the modern science that goes into it. A snapshot of the child's, mother's and alleged father's genetic codes are lined up next to each other in vertical columns, with the child's in the middle column. Since the child inherits from both mother and father, the mother and child's shared markers -- which show up as small horizontal lines -- are crossed out.
That leaves the child and the alleged father. If the child and the man have no markers in common, he is eliminated as a possible father. If the two share a marker, the testers next go to a database that shows how common the marker is in each ethnic or racial group. The rarer the marker, the greater the probability that the child and father are related. The more common it is, the greater the need for additional tests at different genetic sites. The testing center keeps two statisticians on staff to do the math.
For testers who do it every day, the ordinary, inventory-style nature of the business can obscure the all-too-human stories behind the tests. Some of the tests performed at the lab are sweeping social policy writ small. Analytical Genetic Testing, for example, does substantial work for the Immigration and Naturalization Service as part of that agency's attempts to limit the number of people moving into the country.
An immigrant trying to help a relative relocate permanently to the United States has a better shot at succeeding if that relative is a mother or father versus, say, a brother or sister. (There are different standards for different countries; the INS requires a DNA test for Ethiopians merely flying here for a visit.) Many desperate -- or simply less-than-honest -- sponsors are willing to lie about their family ties, and the INS wants to know for sure before it issues another green card.