When DNA testing is done for genealogical purposes, they send back reports with all sorts of incomprehensible information.
I want to save the important information with the person in my genealogy program. I'll create custom fields.
But what I need to know is what data items do I need to create so that I'll have all the DNA info I need in my family file for future analysis?
Many Genealogy programs now have a specific field for this, including Ancestry.com.
The most important thing to record is the most granular Haplogroup that has been determined from your tests.
Also refer to this question on what format to use and field to store it in.
Y-DNA: (Y37, Y67, Y111, Big-Y, etc)
The following is from a YTree Y-Report.
mtDNA:
Autosomal: (AncestryDNA, FamilyFinder, 23andMe)
For Y-DNA and mtDNA, store the raw results that you get, plus the haplogroup.
Y-DNA results are the marker names and their values. ex: DYS393=12
mtDNA results are the position where you differ from the Cambridge Reference Sequence (CRS) plus the base (letter) that you have. ex: 16519C.
For autosomal DNA, you'll need to store the complete raw data file. It's a huge file, so you can't just enter the data into a family tree program.
Some family tree programs now have fields for Y-DNA and mtDNA results, so check to see if your program has them before creating custom fields.
The growing popularity of biochemical genealogy places greater emphasis on the distinctions among data, information and evidence. (See also What is the difference between evidence and information?).
As you have noted, the test provider will return a vast array of data which may or may not be relevant to your current needs (or to your future interests in further investigation). In its raw form, this has little or no value in a genealogical sense. Note also, that the data set you receive will depend upon the choice of tests you had carried out (but more on that later).
Depending upon the test package you have purchased, some of that data will be distilled into information that the provider believes is relevant to your situation. Since most FH software gives you the facility to store {label, value} pairs, you should certainly store that information until you decide what it means to you.
If you have been corresponding with a possible (male) cousin about your (presumed) great grandfather, then two pieces of information about Y-chromosome markers (yours and his) can constitute evidence for the link between your family trees. When you combine the trees, you will cite the DNA test as justification (or, at least, support) for your decision.
At the present time, few family historians understand the type of generic information that is required as evidence for particular problems we face. DNA test providers have tended to concentrate on delivering as much information as possible rather than ensuring they understand what clients actually need.
Would you walk into a vital records registry and ask to purchase certificates for all the Smith births in the nineteenth century? For some of us, getting a DNA test is rather like that. We need to become more sophisticated clients and learn to frame our information requests to match our needs rather than the providers stock.
Begin with the genealogical question that interests you ("Is it true that we have spanish ancestry?") only then can you find out what type of genetic information can provide evidence relevant to that question. Then look within the mass of data provided for information (in this case, on haplogroups) to be recorded as evidence for or against the family legend.
As with most family history searches, the ideal is to understand the question before you start to gather information; although we all have boxes of documents that may or may not be useful. If your aim is to identify a candidate father of an ancestor, then using a mitochonrial DNA test will be a waste of time and money (for that particular question) because the information it will provide cannot constitute evidence. Think of it as being like obtaining a birth certificate in order to identify the cause of death -- all the information is true and concerns the right person, but is useless for the intended purpose.
Every new data type requires us to learn when and how to interrogate it in order to obtain value for our purposes. Biochemical tools are opening up new opportunities, but they will demand study.
A couple of good articles were written by Michele Lewis in the Legacy Family Tree News.
How I use Hashtags to track my DNA matches, Ms. Lewis says if she knows how a DNA tester matches her, she adds their direct line back to their most recent common ancestor into her tree with a special Master Source called "DNA Match - Lineage not confirmed", which will be updated once confirmed genealogically. She then adds a hashtag (a Legacy feature) to the tester and the tester's direct line that says MRCA with the testers who share the MRCA. She does something similar for Y-DNA matches which she describes in the article.
Recording DNA Matches (Intermediate) says to add your "cousin matches" to your genealogy program as unlinked individuals. You can record the company they tested with as an event along with other information, such as kit number and tests done. Ms. Lewis says she only includes the 3rd cousin level people to keep the number of them manageable. She marks them as "invisible" (i.e. research only), so that she never exports that data until she finally links someone up.
What data items do I need to create so that I'll have all the DNA info I need in my family file for future analysis?
There are a couple of basic issues involved when recording data that need to be addressed here.
It isn't always apparent, but most lineage-linked software is designed to record information that we have reached a conclusion about. It isn't the most robust answer when we are recording the "multiple maybes".
It might be more efficient to store your complete DNA results in a separate unstructured journal, and create a note in your lineage-linked software pointing to where the information is in your DNA journal. Your DNA journal could hold
In your family tree software, you could include:
The problem with entering data into genealogy software, in general, is that it encourages us to cherry-pick the information that we think is important at the time we are doing data entry. By nature, we pull this data out of the surrounding context. Even if we make a full transcription of a person's entry in a document like a census enumeration, we generally don't store all the data related to other people in the neighborhood. Because of this, I think relying only on what's in our genealogy program is a mistake. It discourages us from reviewing our original documents as a group and seeing new connections as we find new information.
DNA test results are further complicated because the information we're given in reports is in flux, so rather than storing a single transcription of a document in notes attach to our person, we may need to store a series of snapshots, depending on the analysis of the moment.
So my answer would be:
I would avoid entering information which is volatile. If you want to keep track of how someone's ethnicity estimates have changed over time, that might be better in a separate journal.
