There have been cases (including death penalty cases) in the US where DNA has cemented a guilty verdict, only for it to have later been found to be incorrectly identified/contaminated/planted. DNA is not gospel.
FWIW, I was a geneticist in the University where DNA fingerprinting was invented, and to quote Sir Alec Jeffreys 'DNA creates associations', in that it links the person arising from that DNA to the DNA found at a location. There are various levels of sensitivity and the tests have been refined and improved over the years to provide an astonishing level of accuracy, but there are limits. Some tests are so sensitive that they are liable to cross-contamination, picking up DNA from skin or hair shed naturally or from innocent contact. Finding someone's DNA in their own home would hardly advance a case, but where a bodily fluid is found on or in a victim, or at a crime scene, and the DNA's person and the scene/victim (or victim's DNA on the suspect) are, according to the uncontested prosecution position, not otherwise connected, it makes it rather hard to maintain an alibi or other explanation for the finding, unless of course you have other evidence to indicate that even if the DNA test is accurate, it is not probative of anything.
e.g. if a bent copper gets blood from an alleged drunk driver to test the blood/alcohol level, and splashes a bit on the clothing of a murder victim, then you have an 'association' between the alleged drunk driver and the deceased, who has no explanation for how his DNA was found on the victim other than a supposition that it was planted, but perhaps an alibi that he was in the nick at the time after being caught drunk driving, so bent copper has to be careful and the defence watchful.
And there is the prosecutor's fallacy, e.g. if DNA shows that only 10 people on Earth have the DNA at the crime scene, and Joe Bloggs has that DNA, the chance is not 10/7,000,000,000 or one in seven hundred million that he was not there, but it is one in ten.