Contents

DNA·VIEW NEWSLETTER #16

January 17, 2001

New capabilities in DNA·VIEW

"Roster" import – from Genotyper, Genomyx, or Hitachi StarCall

Import/Export Genotyper import
Hitachi StarCall import
Genomyx M.W. import

The interface with Genotyper and similar files has been refined in very useful ways with DNA·VIEW version 23. Most importantly, you can import the roster. This means, if you already filled in the Genotyper spread-sheet Sample Info column, you won't have to enter the person identification ("accession") numbers again into the DNA·VIEW membrane roster. The roster will be populated automatically, provided the input file conforms to any of several allowable possibilities:
style Sample Info
(or Specimen)
person/patient number 2000-12345
role-within-case (DNA·VIEW style) 16080 C
16080D
16080 f
role-within-case (various acceptable styles) 16080 C2
16080af
etc.

Note: role-within-case requires that the case already be defined.

The way to label quality control samples is to use their DNA·VIEW accession number code, such as 1-562.

No ladder lanes

A membrane is no longer required to have at least one ladder lane. This makes importing (above) a little simpler and more rational.

Read – report format change

The lane report of Read – the screen after digitization showing bands/lane etc. – has been moved to the action menu in order to avoid the annoyance that it could be so big as to interfere with that menu.

Mutation analysis

Mutation history is a new command (DNA·VIEW version 22.37) that analyzes the casework history to estimate mutation rates. This command is extremely useful for labs wanting to conform to AABB rules, or for research.

The output gives both a summary table of apparent mutation rates – abbreviated form:

Locus est mut'ns per 1000 cases with 0 or 1 "exclsn" poss mut'ns
PAC425 Hae 11 1452 16
D3S1358 PCR 1.8 3944 7
vWA PCR 2.8 3627 10
FGA PCR 5.8 3425 20
D8S1179 PCR 3.8 3422 13

and a detail, showing every suspected mutation, e.g.:

(108) Case 99123-10 Residual PI=2,000,000,000 from 13 consistent loci. Inconsistencies:

D3S1358 (b) M= 16 Ch= 18 16 Af= 16 17

(109) Case 99129 Residual PI=100,000,000 from 12 consistent loci. Inconsistencies:

D3S1358 (c) M= 15 17 Ch= 17 15 Af= 16

which gives the information that you need to refine the summary – such as by excluding cases that are really primer dropout, or that are non-paternity, such as uncle cases.

Preliminaries for mutation analysis

Under Paternity Case, Options, choose the switch

Post DNA calculations to a Paradox file?

and be sure it is set to yes. (For PATER users, it will already be.) Once this is done, all cases analyzed by Paternity Case will in the future be available for analysis by Mutation history.

In order to make old cases available for analysis, re-run them. For a moderate number (<100) of old cases, it is reasonably easy using the batch select and batch run options of Paternity case. For a larger number, contact me for special help.

Import Genotypes for database

Import/Export Import genotypes for database (experimental, use at own risk!) is becoming less experimental and more useful. It expects data in columnar format, similar to Genotyper output.

a b c d e f g
Sample Info D3S1358 D3S1358 vWA vWA FGA FGA
WI 1 15 16 14 15 19 21
WI 2 17 18 17 17 19 25
WI 3 16 17 18 18 20 23
WI 4 15 16 15 20 23 26

etc . . .

The user is successively asked to designate (by letter codes, e.g. de) pairs of columns to import as a database. The program makes a guess as to the locus based on the column headings if present, which the user can confirm or correct. The importing process is therefore very quick and painless.

Several databases – corresponding to several loci – imported from the same file will include internal notations so that the DNA·VIEW program Create phenotype list (see below) will be able to reconstruct the multiple-locus genotypes and hence search for duplicates or for a suspect profile.

Database searching and convicted offenders

An old facility, but recently updated for efficiency and for STR's, DNA·VIEW has the capability to maintain and search a convicted offender database.

Protocol for convicted offender database – preliminary

  1. Assign a letter – say z – to code a race anonymous which will be used for anonymous crime samples.
  2. For each locus of interest, use Database Compile a new database from DNA·VIEW data to create a "skeleton" database covering accession numbers from 80-00000 to 9999-00000 and all relevant race codes including z – e.g. cbhz.

Acquire and compile offender and anonymous data

  1. Acquire data for offenders and samples in DNA·VIEW by digitization, by Genotyper import, by Type in a Read, etc. so that each genotype will be associated with a DNA·VIEW accession (=person) number via a membrane roster(s).
  2. Occasionally update/recompile the offender databases (that were initialized in step 2). This is conveniently accomplished using Database, Update an arbitrary list of databases by selecting for the phrase anon. When asked Shall I skip printing the exception reports?, reply y.
Recompiling 100,000-person databases will take 1-10 minutes per locus, and will need 10mb of RAM available to DNA·VIEW (see §XV.B.9.1).

Searching

  1. To search for a particular phenotype (or to search the offender database for duplicates), use Import/Export Create a phenotype list, check duplicates, find multilocus match. The key choices are: The search then proceeds. F5 to print the results.

Miscellaneous

Paternity case output simplified

A few pointless lines are now (ver 23.08) omitted from the Paternity case calculate report output, especially for STR's. The "Gjertson PI" is omitted when it is 0/0; documentation of delta and sigma are omitted when they are 0.

Membrane command bugfix

Membrane   before version 23.11 apparently generated a harmless bug report every 100th attempt to create a membrane.