Well, I just happen to have a book on the shelf with a chapter on that, so:

basically, the chapter (which I can send to you, if you send me a snail mail address) cites no studies of reliability. It simply mentions several times that there is a possibility of laboratory error. It also argues -- reasonably, I would think -- that there are technical reasons why lab error is of less concern for paternity testing than forensic DNA, principally that the stains are plentiful and can be retaken if there is any dispute, neither of which is typically the case for an unknown forensic DNA sample. Also, if a stain were accidentally run twice, this matching itself, lab techs would be alerted since a positive paternity match should not be a complete match, but only a "half" match (since the father supplies 1/2 the genes).

There is a lot of literature, debate over your area, biostatistical issues, in terms of what probability to attach to a match.

Let me know if you want me to send the chapter.

Granted I don't know what the evidence at the trial was, but semen that doesn't match any of the 3 doesn't mean they weren't there.

After a highly publicized trial, Jugum was convicted of second-degree murder, handed a 30-year sentence and banished to Monroe Correctional Complex, effectively disappearing.

Today, Jugum, 56, lives quietly in Bothell, down the street from Pop Keeney Field, in an off-white house with two American flags and at least 50 birdhouses hanging out front.

He's worked at Woodinville Lumber for 25 years, starting out on a prison work-release program and quickly becoming a union shop steward.

He's been married for 24 years to the sister of a fellow Monroe inmate, and he's raised a daughter and son, now 22 and 21.