DNA results that led to Grant conviction 'scientifically corrupt,' says professor

An expert in forensic genetics raised serious concerns about the Thunder Bay laband DNA test results which linked accused killer Mark Grant toCandaceDerksen's 1984 murder in testimony Monday.

"I'm very concerned about the quality of work that came out of that laboratory," BruceBudowlesaid, adding some of the test results were "unexplainable,""scientifically corrupt" and driven by "suspect bias."

Grant was charged with second-degree murder in 2007 in the death of the 13-year-old.Candace'sbody was found frozen and boundwith twine in anElmwoodstorage shed in 1985, seven weeks after she went missing.

• Timeline of theCandaceDerksencase, Winnipeg teen found dead in 1985
• DNA evidence questioned at 1st day of retrial for accused killer ofCandaceDerksen, 13

Budowle, a genetics professor at the University of Northern Texas and former FBI employee, was called as an expert by Grant's defence team to review reports from Molecular World. Police sent several key pieces of evidence to the lab, including the twine, for retesting in 2006.

Police also sent DNA from a new list of persons of interest, including hair and blood samples from Mark Grant, to see if he or others matched the DNA on items seized at the crime scene. The lab waschosenbecause it offered two kinds of DNA tests not previously available to police.

Scientists conducted three rounds of tests. It was DNA on the twine that the lab linked to Grant.

However,Budowlesaid the scientists "incorrectly interpreted" their own findings and ignored data and inconsistencies in the DNA tests that would have excluded Grant.

• RememberingCandaceDerksen: Family braces for retrial of man accused of her murder
• Scientist spoke about 'solving'CandaceDerksencold case before accused went to trial, defence lawyer says

Budowleexplained there was a DNA marker on the twine which could not have come fromCandaceor Grant, but the information was disregarded.

He believes, he told the court, thatinstead ofanalyzingall of the results and then comparing them to Grant's DNA profile, they looked at Grant's sample first and found patterns with the results thendisregarded data which didn't align with their hypothesis.

"The fundamentals of DNA typing were violated in this analysis," he said.

In Budowle's opinion the test results were "at best inconclusive."He also questioned whether the same samples of DNA were used in all three tests.

Grant's defence team also raised questions about the possibility the DNA samples could have become contaminated by lab staff or police investigators who didn't wear full protective gear.

Budowlesaid "significant" contamination was possible.

Crown attorneys Michael Himmelman and Brent Davidson will begin cross-examining him on Tuesday.

The judge-only retrial is being presided over by Justice Karen Simonsen.