According to one source who posted and claims to know:
"Partial or partials?
Multiple contributors?
Yes, yes and yes.
To answer the second question first: Yes there are multiple donors--three--on the tie.
Regarding the first question: All of the DNA related to the three donors is partial.
[Translation: All of the profiles obtained for each donor is a partial profile. ]
Finally, there was one donor who contributed saliva to the tie. It is likely that this is Cooper's DNA given it is easy to envision Cooper smoking, drinking or eating and inadvertently contributing the sample.
None of this explains what happened to the cigarette butts and shaft of hair. Or, why the FBI feels certain enough about their partial DNA profile to publicly exclude Weber and LD by virtue of that DNA while avoiding questions about Sheridan's results."
Who knows how accurate any of this is! Of course we don't know what partial means for any of these profiles? Others go on to claim multiple tests (at least 3) were required for each donor to even obtain a partial in each case! Still SA Carr maintained the profiles they have are good enough to "exclude" people. It's a mess with conflicting stories, but it never prevents Cooperfiles from writing about it ...
There is another guy who states: "The FBI has 14 partials enough to eliminate but not include..." 
So I pose a question for you: How many loci (in Codis-13) do you consider the minimum for a partial? Which loci do you consider most likely to register after 3 runs? Could we assemble our own 'partial model' as a guesstimate of what the FBI must have at minimum?
My understanding is that the above is correct.
It also poses an outstanding question at the end.
I asked the FBI this very question and received the "can't discuss it" response from them. I really do not understand this because they've already stated it's only partial and can exclude some suspects...why not attribute a tangible figure to these statements?
I will say this much, the DNA sample they used for LD Cooper did not come from LD, rather it came from a relative (perhaps daughter). My understanding is that the profile has to be of "reasonable" strength to make that connection. Also, apparently the profile is strong enough for the FBI to actually make the comparisons and exclude Duane and LD.
Two final questions: After taking Sheridan's DNA the FBI investigated him in Nepal and Deer Park, WA...why wouldn't they do this after receiving the DNA results? Did they do this after receiving the DNA results?
How soon after taking Sheridan's DNA did they investigate him in Nepal? I'm assuming you know the sequence of events. Were different agents doing parts of the investigation at different times where one group would get the DNA and one group would do other leg work, or did the DNA sample result in them then moving further along in the investigation?
If the FBI actually has DNA that they think is Cooper's, then they really only need a few of the 13 to rule someone out. So I'm going to hypothesis that they either don't have enough, or they did rule out Peterson and are just not saying it. You could ask the FBI if water is wet and they would give you a standard answer of "we don't discuss that." Somewhere along the line they got bit one too many times and are just not saying anything. Them not answering really means nothing. DNA from a daughter would be plenty to compare.
Georger raises good points. Maybe he has some contacts who could help determine what partial would be needed. I wonder about the DNA. Hypothetically they could have samples from 3 different people on the plane, but that does not mean that all 3 samples are from the same genetic markers, so you'd have to rule out a suspect for all 3 of those samples, which means if the samples are partial, a suspect could maybe be ruled out for 2 of the 3, leaving him questionable for the 3rd. As some of the legal experts have said here before, the government needs a conviction, and therefore needs a great match, maybe even at 13 markers.
I'm not a legal expert or a DNA expert, but I've studied the science and the probabilities, and to be a match for even 3 or 4 of the 13 is highly improbable. The government would not put you in jail for murder on just 3 or 4 of the 13, but if there was other evidence, it would help the case. So if the FBI has the DNA, and they have 3 or 4 markers, and then they find someone who fits the profile (description, experience, grudge, etc.), then that should be enough for them to say they have their guy. A conviction is next to impossible at this time I bet.
DNA is a 13 digit lottery number, in order, with decimals within those 13. The odds of a match are in the trillions. If the FBI has DNA, then they can rule someone out easily. I just don't think they have what they need.