The honest scientist recognizes that she herself is a test instrument, and a fallible one at that. Subjectivity inescapably enters into any human endeavor, and should not be denied. DNA testing is rife with subjective elements, no place more so than at the crucial stage of deciding whether a match exists. On the one hand, non-matching extraneous bands may sometimes be properly disregarded and patterns that do not quite meet objective matching criteria may be appropriately regarded as incriminatory matches. On the other hand, band patterns that do meet objective matching criteria may be treated as exonerative depending on how they deviate from perfect matches. The DNA expert should not hide behind the cloak of science to deny the role of human judgement. White coats should not be worn into the courtroom either literally or figuratively.
At the same time, the honest scientist tries to be as objective as possible in her judgements. She realizes that this is inconsistent with a strong a priori belief that the donor of a suspect sample is guilty. Thus she avoids any information suggesting the involvement or uninvolvement of the accused until after she has prepared her report and, in the ideal case, until after she has testified. Laboratories should cooperate to make this easy. Crime-related information should be stripped from all information sent to the analyst unless it is essential to the test and its interpretation (e.g. information that severn people are suspected of participating in a rape or that the suspected rapist is the victim's brother).
Lempert, Richard O. "The Honest Scientist's Guide to DNA Evidence." Genetica 96, no. 2 (1995): 119-124.