Building a Database in DNA·VIEW

Overview

At any time, DNA·VIEW can be asked to compile an allele frequency database, based on accumulated profiles.

Take a look at VII.A.3 so that you have an idea of how the program will view your data. The genotypes of anyone with an accession number can be included in the database. That certainly includes anyone in a Case (Paternity, Kinship, etc.), but more generally anyone who occurs on a (membrane) roster even if they are not used in any Case.

Any single database pertains to

• one specified locus
• one or more specified race letters.
  That is, people are chosen for inclusion according to the race letter that is part of the definition of that person. In fact, selection for inclusion in a database is the whole reason for associating a race letter with a person. (The specification of the race to be used for calculation comes from a race letter associated with a Case, or sometimes chosen by the user in response to a prompt.)
• a range of accession (person) numbers.
  most often I select the full range of accession numbers – from 80-00000 to 9999-99999, but if perchance you have in mind to create many categories for population studies, advance planning of your accession number scheme may be helpful.
• a set of "Reader numbers"
  The default, usually most useful is to use all of them (099=Genotyper might be the only one), except for 000=Student.

Preparation

Routine

So keep the above in mind while designing data encoding and calculating cases. In particular, you probably want to reserve particular letters for the likely races of interest – e.g. p for Puerto Rican. Then, every man and every woman (but preferably not the children) should have the letter attached to their person-description.

There are a few ways of doing this. You may want to use a combination.

Assign race to an entire worklist

With Assign race to membrane (Disaster Analysis menu)
• select a collection of membranes/worklists;
• say whether to exclude children (probably yes);
• choose some race (letter) such as p.
Then all the denominated people will be assigned to the desired race.

Assign race to people in a case

For each case, select the case (in Paternity case), choose edit people, and the editing cursor will appear on the first person, the mother.

• space twice to go past the accession number
• type e.g. p in the race cell
• tab to skip to the next person
• tab to skip past a child
• fix the (alleged) father(s) in the same way as the mother.

The most efficient practice is to make a habit of doing this while you are computing the cases. Then the data is ready to create databases whenever you like.

Assign race to individual people in a worklist

• Open the worklist with Membrane from the Casework menu.
• Select the lane with the person of interest.
• Space throught the fields to edit the person an introduce the desired race.

Preliminary check

The Case List option (in Paternity Case) can be used to quickly review a sequence of cases to check if you have properly filled in the race letter for each person.

Compile database

In fact, I wouldn't bother to wait for 100 cases, because once you make them the first time to REbuild is extremely quick easy – just one operation for all 13 loci. See VII.A.5.c.

Initial creation of databases

Even with no cases at all, you can build the (empty) databases for your lab. When you do it the first time, there is a separate operation for each locus, so you will need 2 or 3 minutes to do all 13 (or so) loci. The procedure is straightforward:

Step-by-step

• Populations
• Database
• Compile a new database from DNA·VIEW data
  • Which locus?. Select the desired locus.
  • Which readers?. Enter for default choice(s).
  • Which race(s)?. Type 1 to 6 letters designating races to be include in the database, Enter
  • Earliest acceptable accession ID 80-0000. Enter to accept the default
  • Last acceptable accession ID 9999-0000. Enter to accept the default
  • (time passes, probably a few seconds, and counters indicate progress as the program finds and compiles the relevant data)
  • Select the database to update, or End to make a new one and the list of already existing databases is displayed. Usually, End to add a new database to the list.
  • Esc to proceed to the next locus.

Updating the databases

Once the databases have been built the first time, they can easily later be updated to reflect additional people who have been entered into DNA·VIEW.

There are two easy ways to do this. The first method assumes that you have marked as defaults the databases that you want to update. If you've already been using them for casework, then this would be the case. Or, if you're planning to use them for casework after the re-compilation, then go ahead and mark them as defaults now, the proceed with re-compilation.

Step-by-step (re-compile default databases) Populations Database Update the default databases (but only those marked as from your own laboratory)

The second method is more general, but takes (just) a few more keystrokes:

Step-by-step (recompile arbitrary list of databases) Populations Database Update arbitrary list of databases Select databases to update Accept lines containing My lab Enter to limit the list to "My lab"'s databases. Select databases to update Accept lines containing If desired to restrict the list further, type a context then Enter Select databases to update Accept lines containing Enter to proceed with the compilation. Obtain and drink coffee

---