A Using GeneScan with the ABI 373. A-1
Introduction. In This Appendix. Processing Using the ABI 373. Introduction. ABI 373 Versus ABI PRISM 377 Data. Using the GeneScan Analysis Software Before Version 2.0. Using the GeneScan Analysis Software Version 2.0 or Later. Processing Steps. Comparison of GeneScan Analysis. Setting Gel Processing Parameters. A-1 A-1 A-2 A-2 A-2 A-2 A-2 A-3 A-3 A-5
Introduction. A-5 Procedure. A-5 Opening the Collection File. A-7 Introduction. A-7 Procedure. A-7 Completing the Sample Sheet. A-9 Introduction. A-9 Procedure. A-9 Tracking Lanes. A-11 Introduction. A-11 Process. A-11 Procedure. A-11 Generating Sample Files. A-12 When to Generate Sample Files. A-12
B GeneScan Size Standards. B-1
Introduction. B-1 In This Appendix. B-1 GeneScan 350 Size Standard. B-2 What To Use It For. B-2 How It Is Prepared. B-2 GeneScan 350 Molecular Lengths. B-2 Running Under Denaturing Conditions. B-2 Double-Stranded GeneScan 500 Fragments. B-3 GeneScan-400HD Size Standard. B-4 What To Use It For. B-4 Special Uses. B-4 How It Is Prepared. B-4 Fragment Lengths. B-4 Denaturing Electropherogram. B-5 Non-denaturing Electropherogram. B-5 GeneScan 500 Size Standard. B-6 What To It Use For. B-6 How It Is Prepared. B-6
GeneScan 500 Molecular Lengths.B-6 Running Under Denaturing Conditions.B-6 Double-Stranded GeneScan 500 Fragments.B-7 GeneScan 1000 Standard.B-8 How It Is Prepared.B-8 GeneScan 1000 Molecular Lengths.B-8 Running Under Native Conditions.B-9 GeneScan 2500 Standard.B-10 How It Is Prepared.B-10 What To Use It For.B-10 Running Under Native Conditions.B-10 GeneScan 2500 Molecular Lengths.B-11
C Size Calling Methods. C-1
Introduction.C-1 In This Appendix.C-1 Least Square Method.C-2 About This Method.C-2 Advantages.C-2 Cubic Spline Interpolation Method.C-4 About This Method.C-4 Possible Local Sizing Inaccuracy.C-4 Local Southern Method.C-5 About This Method.C-5 The Equation.C-5 How This Method Works.C-6 Global Southern Method.C-7 About This Method.C-7 The Equations.C-7 How This Method Works.C-8
D Troubleshooting the GeneScan Software. D-1

Introduction. D-1

In This Appendix. D-1 Troubleshooting Projects and Results. D-2 Table Description. D-2 Troubleshooting Gel Data. D-5 Table Description. D-5 Troubleshooting Genotyping Results. D-8 Table Description. D-8 GeneScan Error Messages. D-9 Introduction. D-9 GeneScan Analysis Software Crashes with BioLIMS. D-9 Analysis Log Error Messages. D-10 Error Messages When Dening Size Standards. D-11
E Using the BioLIMS Database. E-1
Introduction. E-1 In This Appendix. E-1 About the GeneScan Analysis Software and the BioLIMS Database. E-2 What is the BioLIMS Database?. E-2 Accessing the BioLIMS Database. E-2 Before Using the BioLIMS Database. E-2 Modes in GeneScan Analysis Software. E-2 Comparing Modes. E-3 Conguring the BioLIMS Database Server. E-4 Sybase or Oracle?. E-4 Conguring for Sybase SQL Server Connection. E-4 Conguring for Oracle Server Connection. E-7 Switching Between Sample File and BioLIMS Mode. E-10 Introduction. E-10 Switching Modes. E-10 How to Access the BioLIMS Database. E-12 Introduction. E-12 Before Accessing a Sybase Database. E-12 Accessing the BioLIMS Database. E-13 About Server Names. E-17

c. d. by phone from outside the United States or Canada a.
Press 1 to order an index of available documents and have it faxed to you. Each document in the index has an ID number. (Use this as your order number in step d below.) Call 1-650-596-4419 from a touch-tone phone a second time. Press 2 to order up to ve documents and have them faxed to you.
1-22 GeneScan Analysis Software Overview
To Reach Us by Contact technical support by e-mail for help in the following product E-Mail areas.
For this product area Chemiluminescence Genetic Analysis LC/MS PCR and Sequence Detection Protein Sequencing, Peptide and DNA Synthesis Use this e-mail address info@tropix.com galab@perkin-elmer.com apisupport@sciex.com pcrlab@perkin-elmer.com corelab@perkin-elmer.com
Regional Ofces If you are outside the United States and Canada, you should contact Sales and Service your local PE Applied Biosystems service representative.
The Americas United States PE Applied Biosystems 850 Lincoln Centre Drive Foster City, California 94404 Tel: Fax: (650) 570-6667 (800) 345-5224 (650) 572-2743 Canada (Mississauga, Ontario) Tel: Fax: (905) 821-8183 (800) 668-6913 (905) 821-8246
Latin America (Del.A. Obregon, Mexico) Tel: Fax: (305) 670-5310 (305) 670-4349
Europe Austria (Wien) Tel: Fax: Hungary (Budapest) Tel: Fax: 14 61
Benelux (Nieuwerkerk a/d IJssel, Netherlands) Tel: Fax: 31(0) 31(0) 180 331409
Italy (Milano) Tel: Fax: (039) 23831 (039) 2383490
Chekia Rep. (Praha) Tel: Fax: 21 68
Norway (Oslo) Tel: Fax: (0) 1500 (0) 1501
GeneScan Analysis Software Overview 1-23
Europe (continued) Denmark (Allerd) Tel: Fax: Poland (Warszawa) Tel: Fax: (48 22) 10 (48 22) 20
Finland (Espoo) Tel: Fax: 72 701
Russia (Moskva) Tel: Fax: 564 8787
France (Paris) Tel: Fax: (1) (1) 85 00
South Africa (Johannesburg) Tel: Fax: 478 0349
Germany (Weiterstadt) Tel: Fax: (0) 6150/ (0) 6150/ 101 101
Spain (Madrid) Tel: Fax: (1) (1) 804 0414
Sweden (Sundbyberg) Tel: Fax: (0)4400 (0)4401
United Kingdom (Warrington, Cheshire) Tel: Fax: (01925) 825650 (01925) 282502
Switzerland (Rotkreuz) Tel: Fax: (0) 7708 (0) 0676

Sequencer instrument, and the ABI PRISM 377 with XL upgrade, stores the raw data collected during the entire run of the instrument. A gel le can be 10 MB to 60 MB.
96-Lane Gels The GeneScan Analysis Software can open and analyze 96-lane gels. Capability Gel File Contents Initially, the le contains the following:
o o o o Raw data collected during the run. Gel image, which is similar to an autoradiogram, but in color. Copy of the Data Collection software sample sheet. Copy of the matrix le.
After lane tracking and editing, the le also contains the lane tracking information and any changes you make to the original information in the le.
How GeneScan When the GeneScan Analysis Software tracks the gel le, it creates a Tracks a Gel File tracker line for each lane in the gel. Each tracker line is trying to nd the
center of the bands, not the brightest part of the band. The tracker will only track if a matrix is attached to the gel le. During data extraction, the software generates a Sample le for each tracked lane by averaging the data from the tracked channel and the number of channels you specify on either side of it. The software also copies to each new Sample le all the information required to identify and analyze the sample.
2-2 How to Process and Edit the Gel File
How to Set Gel Processing Preferences
Introduction Use the Gel Processing dialog box to dene the parameter values to
use for gel le processing. The following procedure applies to the ABI 373 and the ABI PRISM 377 instruments.
Displaying the Gel To display the Gel Preferences dialog box: Preferences Dialog Step Action Box
If the GeneScan Analysis Software is not running, then double-click the program icon. Choose Gel Preferences from the Settings menu. The Gel Preferences dialog box appears.
There are three Gel Preferences parameters: o o o Auto-Launch Processing. Image Generation Defaults. Lane Extraction.
How to Process and Edit the Gel File 2-3
Auto-Launch Select or de-select the checkboxes to specify whether or not the gel is Processing to be automatically tracked and extracted at the end of a data collection
Select both checkboxes to analyze samples automatically after data-collection.
Image Generation In the Image Generation Defaults options enter values that determine Defaults how the GeneScan Analysis Software processes the gel le when it
generates the initial gel image from data collected after electrophoresis. Image Generation options:
Item Scan Range Description Enter the scan numbers at which you want to display the gel image. Use a trial run to determine the scan range since the time frame in which fragments are detected may vary from run to run. During analysis, the GeneScan Analysis Software ignores all scans outside the specied scan range, and only includes the scans with the designated ranges when it generates Sample les. See Regenerating the Gel Image on page 2-23 to change the number of scans displayed.

The Neural Net Tracker uses special tracker setting les that are optimized according to the number of channels and lanes in the gel le and the comb type. It is important that you set the correct comb type for the gel so that the Tracker applies the correct tracker setting les. This value appears at the bottom of the gel window (see Gel File Window Diagram on page 2-13).
IMPORTANT The proper comb type needs to be selected in order to ensure proper tracker performance.
2-10 How to Process and Edit the Gel File
How to Display the Gel File
Introduction Use the gel image to observe sample migration, lane tracking, and
signal strength in the gel image. You can also use the window to adjust the image to improve visibility, re-align individual lane markers, or edit the position of the tracker lines.
Displaying the Gel You can display a gel le as follows: File
To display a gel le. automatically Then. after automatic analysis, the Gel File window opens and displays the newly created gel le. For more information, see Setting Up for Automatic Analysis on page 4-4. manually you can either: Double-click the Gel le, or a. b. c. Choose Open from the File menu. The Open Existing dialog box appears. Click the Collection Gel icon. A directory dialog box appears. Locate and select the desired gel le and choose Open.
How to Process and Edit the Gel File 2-11
About the Gel File Window
Differences The following table explains the differences between the GeneScan Between Gel File Analysis Software Gel File window and the Gel File window displayed Windows by the Data Collection software during a run.
Application Data Collection software Gel window Shows Real-time data as it is being collected. New Data New data appears at the bottom of the screen as it is collected, so the top of the screen shows the start of the run. This image is inverted. The bottom of the window displays the start of the run, that is, the smallest fragments appear at the bottom of the window, just as they would on an autoradiogram. Gel Image Gel image is not baselined or multicomponented

GeneScan Gel File window

An image of the gel after the Data Collection software is nished.
Gel image is baselined and may or may not be multicomponented.
2-12 How to Process and Edit the Gel File
Gel File Window The following is an example of the Gel File window. Diagram

Saving Files After IMPORTANT If you save changes to the gel le, the original tracking Editing Tracking information is overwritten. You can retrieve the originally calculated tracking only
by choosing Track or Track & Extract Gel from the Gel menu to retrack the gel. See Tracking Lanes and Extracting Data on page 2-59.
You have the following options:
If you. choose the Save command select the checkbox labeled Save Gel after Extraction in the Extract Lane dialog box Then. The contents of the Gel File window are saved to the gel le. It is not necessary to use the Save command to save the gel le.
For information regarding saving selected portions of gel les and archiving gel les, see Saving Gel Files on page 8-4.
2-68 How to Process and Edit the Gel File

Analyzing Sample Files

In This Chapter Topics in this chapter include the following:
Topics About Sample Files Opening Sample Files About The Sample File Window Sample Results View Sample Info View Size Curve View Raw Data View EPT Data View Analyzing a Sample File
See page 3-2 3-4 3-5 3-6 3-8 3-13 3-15 3-17 3-19
Analyzing Sample Files 3-1

About Sample Files

What Sample Files Sample les contain data generated from the following: Contain
Generated from. lane information from gel les Using. ABI 373 or ABI PRISM 377 instruments. After extracting lanes from a gel le, Sample le data can be analyzed as follows: Analyze Sample le data. directly from the Sample le from the Analysis Control window in a project
For information, see. Analyzing a Sample File on page 3-19. Analyzing Sample Files: Using the Analysis Control Window on page 5-6.
capillary electrophoresis
ABI PRISM 310 instrument.
How GeneScan The GeneScan Analysis Software creates Sample les after extracting Generates Sample lanes. The information from each lane in the gel le is tracked and Files extracted, and the resulting Sample les are placed in their respective
sample folder. If you change tracking, lane assignment, or Sample Sheet information, you have to regenerate the Sample les. The software consults Sample Sheet information to determine whether a lane is used (contains sample). The lane tracker uses this information to assign lane numbers to the tracker lines. In addition, the GeneScan Analysis Software only extracts those lanes identied as Used.
3-2 Analyzing Sample Files
Ways to Generate There are two ways to generate sample les: Sample Files

Software Data Collection software GeneScan Analysis Software Instrument ABI PRISM 310 ABI 373 or ABI PRISM 377 Generates Sample le for each injection. Sample le for each lane.
How GeneScan The GeneScan Analysis Software performs the following steps in Analyzes Sample analyzing Sample les: Files
Step 1 Action Processes the raw data signals to generate analyzed data signal and then uses the analyzed signals to detect the signal peaks associated with DNA fragments. Performs size calling by identifying the peaks of the in-lane size standard found in each sample. Determines the fragment size of each experimental peak within the sample based on the size calling curve generated using the size standard peaks, the selected size calling method, and by comparing it to the pre-dened size standard le. The algorithmic steps of the process from raw data to analyzed data are as follows: o o o o Multicomponenting. Baselining. Smoothing, if any. Peak detection.
Analyzing Sample Files 3-3

Opening Sample Files

Introduction Sample les can be opened as separate les outside of projects, and
display related information about each Sample le.
If you are interested in. one or two Sample les Then. it is often more convenient to open Sample les individually and analyze or view the data without or opening an entire project. use a project.

multiple Sample les

For information on opening Sample les from within: o o Projects, see Accessing Sample Files on page 5-6. BioLIMS, see Switching Between Sample File and BioLIMS Mode on page E-10.
Procedure To open a Sample le as a separate le:
Step 1 Action Choose Open from the File menu. The Open Existing dialog box appears. Note You can also double-click the Sample le name in the Finder. If the GeneScan Analysis Software is not running, the software starts and opens the Sample le.
Click the Sample icon. A directory dialog box appears. In the dialog box, nd and select the Sample le that you want to open. Click Open. The Sample File window appears. See About The Sample File Window on page 3-5.
3-4 Analyzing Sample Files
About The Sample File Window
What it Displays You can use the different display modes in the Sample File window to
review the analyzed and raw data, and all pertinent data collection, sizing and Sample description information from a single window. The Sample Results view appears as the default.
Five Views The ve views of the Sample File window are:
View Sample Results view Sample Info view Size Curve view Raw Data view EPT Data view See page 3-6 3-8 3-13 3-15 3-17

Analyzing Sample Files 3-5

Sample Results View

What it Displays The Sample Results View displays the Sample les analyzed data in
both electropherogram and tabular data form.
Displaying the The following table lists ways to display the Sample Results View: View
To display the view. from the Sample le window Do this. o Click the button for the Sample Results view at the bottom left of the Sample le window, or, Choose Sample Results (z E) from the Sample menu. Select a sample or multiple samples in the Analysis Control or the Results Control window. Choose Sample Results from the Sample menu.
o from a project window a.
Sample Results The following is an example of the Sample Results view: View Example
3-6 Analyzing Sample Files
Description of The following table describes the columns in the above gure: Columns
This column Dye/Sample Peak Identies o o Minutes Size Dye color. Peak number.
The time, in minutes, from the start of the run to the time the fragment was detected. The number of base pairs in the fragment. This value is calculated automatically only if you: o o Run the size standard in the same lane or injection as the sample, and Perform size calling.
Peak Height Peak Area Data point
Signal size. Area of the detected peak. Data point of the fragment at its maximum peak height. Scan number for ABI 373, ABI 373XL, ABI PRISM 377 and ABI PRISM 377XL data.
Differences From The Sample Results view displays the same electropherogram and the Results Display tabular data as the Results Display, with the following differences:
o o o o One Sample le displayed. Show or hide dye/sample data by clicking the buttons below the electropherogram. Cannot display legends. Cannot use custom plot colors.
Analyzing Sample Files 3-7

Sample Info View

What it Displays Displays the following Sample le information:
o o o o Run and data collection information. Gel information. Sample informationthis information can be edited. Analysis records.
Displaying the The following table lists ways to display the Sample Info View: View
To display the view. from the Sample le window Do this. o Click the button for the Sample Info view at the bottom left of the Sample le window, or, Choose Sample Info (z I) from the Sample menu. Select a sample or multiple samples in the Analysis Control or the Results Control window. Choose Sample Info from the Sample menu.
The information is organized in ve panels. Click the triangles to expand or collapse the panels to display specic information.
3-8 Analyzing Sample Files
Sample Info View The following is an example of the Sample File window in Sample Info Example View:
Description of The following tables list the information in the Sample Info View: Information

height to be detected for analysis. This, in turn, controls the number of peaks analyzed. Peaks falling below the parameters specied appear in the electropherogram, but are not analyzed, and no values appear for them in the tabular data. Peak Detection Parameter Options Described
Item Dye Amplitude Threshold Description Set the dye amplitude threshold at a level that allows the software to detect peaks, but eliminate noise. For each dye, the GeneScan Analysis Software detects peaks above the threshold entered in the entry eld. For example If you leave the default value of 50, peaks with amplitude above 50 are analyzed and appear in the tabular data. Lower amplitude peaks still appear in the electropherogram, but are not analyzed and do not appear in the tabular data.
5-20 Analyzing Project Files
Item Minimum Peak Half Width
Description Denes what constitutes a peak. Use to specify the smallest half peak width for peak detection. The range is from 2 - 99. A typical number might be 3 for microsatellites, or 10 for SSCPs.
For example If this number is large, the software ignores noise spikes. If the peaks in the data are narrow, set the value to a low number. Experiment with this value to determine the best number for the data.
Size Call Range About the Size Call Range Parameter Options Parameter Options Use the Size Call Range parameter options to specify the range of size
fragment (in base pairs) to be included in the peak tabular data. Size Call Range Parameter Options Described
Item All Sizes radio button This Range (Base Pairs) radio button Description All detected fragments appear in the tabular data. Dene lower (minimum) and upper (maximum) limits of the peaks to include in the tabular data.
Size Calling About Size Calling Method Parameter Options Method Parameter Click a radio button to select the desired size calling method. The Options GeneScan Analysis Software uses these methods to determine the
molecular length of an unknown fragment. Size Calling Method Parameter Options Described
Item 2nd Order Least Squares and 3rd Order Least Squares Description Both Least Squares methods use regression analysis to build a best-t size calling curve. For more information, see Least Square Method on page C-2.
Analyzing Project Files 5-21
Item Cubic Spline Interpolation
Description Forces the sizing curve through all the known points of the selected GeneScan size standard. For more information, see Cubic Spline Interpolation Method on page C-4.

Local Southern method

Determines the sizes of fragments by using the reciprocal relationship between fragment length and mobility. For more information, see Local Southern Method on page C-5.

Examine the display to ensure that each peaks center, beginning, and end points are correct. Choose Hide Peak Positions from the View menu to suppress the display of the peak markers. Note You can also use the Sample Info view to display information about the peaks detected and matched. For more information, see Description of Information on page 3-9.
Evaluating Analysis Results 7-53
Saving, Archiving and Copying Files 8
In This Section Topics in this chapter include the following:
Topics Saving GeneScan File How to Archive Sample and Gel Files Transferring Data to Other Applications

See page 8-2 8-6 8-7

Saving, Archiving and Copying Files 8-1

Saving GeneScan File

Why Save The following table lists why to save projects, Sample les, and gel les.
For information on archiving les, see How to Archive Sample and Gel Files on page 8-6.
Save GeneScan projects Because Protects the links to Sample les and their preferences. Projects contain links to Sample les and preferences regarding display and analysis. Sample les Protects the links to projects and their preferences. Sample les also contain raw data and critical information about the run, settings, and analysis control. Gel les Preserves the integrity of the data. Note If you require long-term storage of multiple gel les, saving selected information from the les reduces their size considerably. Results Displays Saves the Results Display settings in projects when you have found a display format that suits your needs. Saving Results Displays on page 8-5. Saving Gel Files on page 8-4. Saving Sample Files on page 8-3. See Saving Projects on page 8-3.
8-2 Saving, Archiving and Copying Files

Saving Projects Note

You do not need to save a Sample le after analysis. The analyzed data is written directly to the Sample le during analysis.
The following table lists the options for saving a project:
If you choose. Save Project (z S) Then. you can take the following action: If you. previously saved the project had not saved the project Then. it is automatically saved using the same name. a dialog box appears. Select a location for the le, enter a name, and click Save.

Save Project As

a dialog box appears. Select a location for the le, enter a name, and click Save.
Saving Sample Choose Save Sample Info (z S) from the File menu. Files Note If you choose Close from the File menu, or click the close box when

GeneScan 1000 Fluorescently labeled native fragments are 18 nucleotides longer than Molecular Lengths denatured fragments (see the table below). You can use this standard to
size fragments in the 100 to 900 base pair range. GeneScan 1000 Fragment Molecular Lengths (Base Pairs):
Unlabeled 29 Native (+36) 47
B-8 GeneScan Size Standards
Running Under The following diagram shows the peak pattern of fragments run under Native Conditions native conditions on an ABI PRISM 310 using 2.5% GeneScan Polymer
Solution in a 30 cm Ld capillary.
IMPORTANT An asterisk (*) for the 262 and 692 base pair peaks denotes peaks resulting from abnormal migration of double strands that did not completely separate under denaturing conditions when analyzed on the ABI PRISM 310. Do not use these peaks to size samples. The peaks show smaller values than the actual size of the fragments. Refer to the GeneScan Reference Guide, Chemistry Reference for the ABI PRISM 310 Genetic Analyzer (P/N 4303189 rev A) for further details. Asterisks
GeneScan Size Standards B-9

GeneScan 2500 Standard

How It Is Prepared The GeneScan 2500 standard is made from lambda phage DNA
restriction digested with Pst I, followed by ligation of either a TAMRA or ROX oligonucleotide. It has 28 fragments, ranging from 55 to 14, 097 base pairs (bp).
What To Use It For You can use the GeneScan 2500 standard for native applications to size
fragments in the 100 to 5,000 base pair range.
Running Under The following gure shows the peak pattern of fragments run under Native Conditions native conditions on the ABI PRISM 310 using 2.5% GeneScan Polymer
IMPORTANT An asterisk (*) for the 508 base pair peaks denotes peaks resulting from abnormal migration of double strands that did not completely separate under denaturing conditions when analyzed on the ABI PRISM 310. Do not use these peaks to size samples. The peaks show smaller values than the actual size of the fragments. Refer to the GeneScan Reference Guide, Chemistry Reference for the ABI PRISM 310 Genetic Analyzer (P/N 4303189 rev A) for further details.

Use the pop-up menu to add, change, or remove aliases. If you have more than one alias, click the Make Default check box to choose which one appears when you rst open the Edit Session dialog box.
Note The default alias is the database that opens if you choose to automatically analyze data. Note If both the Make Default and the Save Password boxes are checked, no dialog box will appear when a connection to the server is requested. Since all the information required of the user has been saved, the software will connect to the database automatically.
Using the BioLIMS Database E-15
Step 9 Action Take one of the following actions: If the login was. successful Then. a. Choose Open from the File menu. The Open Existing dialog box appears. b. Click the Sample icon and the Collection Browser appears.
For more information, see Using the Collection Browser Window on page E-20. unsuccessful an alert dialog box appears. Check that: o o o All the login information was entered correctly and in the correct case. Your interfaces le is correctly congured (page E-4). If the connection is still not open, consult Appendix F, Troubleshooting the BioLIMS Database.
E-16 Using the BioLIMS Database

About Server Names

Sybase or Oracle? The BioLIMS Session Manager decides whether you are connected to
Sybase SQL Server or to an Oracle Server database by looking at the name in the Server eld in the Session Manager dialog box.
How Names are The table below summarizes how names are recognized. Recognized
If the Session Manager sees a Server name All in uppercase letters Sufxed by :s or :S Containing any lowercase letters Sufxed by :o or :O
It assumes a Sybase SQL Server database connection. Sybase SQL Server database connection. Oracle Server database. Oracle Server database.
Example MOZART Offenbach:S Oramozart SIBELIUS:O
Sybase SQL Example 1 Server Examples If the interfaces le contains this:
MOZART query MacTCP mac_ether mozart.apldbio.com 2500 MOZART is recognized as a Sybase SQL Server because the server name is in all uppercase letters. The Session Manager would look like this:
Using the BioLIMS Database E-17
Example 2 If the interfaces le contains this: Offenbach query MacTCP mac_ether mozart.apldbio.com 2500 The Session Manager would look like this:
In order for Offenbach to be recognized as a Sybase SQL Server, the name in the Server eld is sufxed with :S.

y-axis positions, displaying in electropherograms 7-27
zooming, electropherograms 7-31
unlocking sample les 4-12
vertical scale, changing 7-38 to 7-39 for all electropherograms 7-38 for single electropherograms 7-39

Index-7

Arch Dis Child Fetal Neonatal Ed 1999;81:F159F160 The results of genotyping analysis of the BUGT1 gene in 55 healthy infants are given in table 1. The one variant homozygote was omitted from the statistical analysis. The serum bilirubin concentration at 4 days of age in this infant was 9.1 mg/dl. In the healthy infants no signicant diVerence was detected in serum bilirubin concentrations at 4 to 5 days of age between normal homozygotes (10.0 (2.7) mg/dl; mean (SD)) and heterozygotes (9.2 (1.5) mg/dl) (p = 0.43, unpaired Students t test). We also analysed 19 infants with jaundice requiring treatment; 18 normal homozygotes and one heterozygote. Thus the TA-7 allele was found in only one of 19 cases. The peak serum bilirubin concentrations in the 18 normal homozygotes were 18.8 (2.29) mg/dl and that in the heterozygote was 15.7 mg/dl. TA-7 allele frequency was calculated to be 0.07, signicantly lower than the value of 0.4 found in the North American and Eastern Scottish populations (p < 0.001, 2 analysis with one degree of freedom).The genotype distribution in the 74 Japanese infants was also signicantly diVerent from that found in the North American and Eastern Scottish populations (p < 0.001, 2 analysis with two degrees of freedom). Ethnic diVerences in the incidence of neonatal jaundice have been reported. Neonatal jaundice occurs more often in East Asian infants than in Caucasian infants.46 Even if the presence of TA-7 could aVect the metabolism of bilirubin in the neonatal period, it does not explain the high incidence of neonatal jaundice in Japanese infants, because the TA-7 allele frequency is very rare in the Japanese population. In the 74 infants in this study, we detected only one who was homozygous for TA-7, which happened to be a baby who was in the healthy control group. In the 19 infants with jaundice requiring treatment, we found a TA-7 allele in only one heterozygous case. In conclusion, our ndings indicate that the variant TATA box in the promoter region of the BUGT1 does not contribute to the high incidence of neonatal jaundice in the Japanese population.
SHOZO WAKU YASUHIRO TAKESHIMA HAJIME NAKAMURA Department of Paediatrics Kobe University School of Medicine 7-5-1 Kusunoki-cho Chuo-ku, Kobe 650-0017 Japan HISAHIDE NISHIO KIMIAKI SUMINO Department of Public Health 1 Bancroft JD, Kreamer B, Gourley GR. Gilbert syndrome accelerates development of neonatal jaundice. J Pediatr 1998;132:656-60. 2 Bosma PJ, Chowdhury JR, Bakker C, et al. The genetic basis of the reduced expression of bilirubin UDP-glucuronosyltransferase 1 in Gilberts syndrome. N Engl J Med 1995; 333:1171-5.

LETTERS TO THE EDITOR

A variant TATA box in the bilirubin UDP-glucuronosyltransferase 1 gene promoter does not contribute to neonatal jaundice in the Japanese population EDITOR,A variant TATA box (A[TA]7TAA, designated as TA-7) in the promoter region of the bilirubin UDP-glucuronosyltransferase 1 (BUGT1) gene has been reported to accelerate the increase in bilirubin concentration during the rst two days of life.1 However, the relation between TA-7 and peak bilirubin concentration in the neonatal period has not been claried, and it is also unclear as to whether TA-7 inuences bilirubin metabolism in infants of diVerent ethnic groups. To investigate whether TA-7 is one of the risk factors for neonatal jaundice in the Japanese population, we performed genotyping analysis of the BUGT1 gene in 74 Japanese newborn infants, and measured bilirubin concentrations at 45 days of age in healthy infants (n=55) and peak bilirubin concentrations in infants with jaundice requiring treatment, phototherapy, or exchange transfusion (n=19). Informed consent was obtained from parents, and infants were enrolled at birth. All infants were born at 3742 weeks of gestation and weighed more than 2500 g. They showed no blood incompatibility, had a negative direct Coombs test; they had no clinically signicant antenatal and intrapartum complications, and no clinically detectable pathology except for jaundice. In healthy infants serum bilirubin concentrations were measured at 45 days after birth. In infants with jaundice requiring treatment, peak bilirubin concentrations were used in the analysis. The TATA box region of the BUGT1 promoter was amplied with the polymerase chain reaction (PCR) using the primers BUGT1-A (5'-TAACTTGGTGTATCGATTGGTTTTTG-3'1) and BUGT1-B(5-ACAGCCATGGCGCCTTTGCT-3'1). BUGT1-B was labelled with a uorescent dye, 6-carboxyX-rhodamine. An aliquot of each PCR product was electrophoresed on a genetic analyser (ABI PRISM 310; Perkin-Elmer Applied Biosystems, Foster City, CA, USA) and then analysed using the GeneScan analysis software package (Perkin-Elmer Applied Biosystems). The electrophoretogram clearly separated the peaks for a normal TATA box (A[TA]6TAA, designated as TA-6, 90 base pairs of PCR product size) and TA-7 (92 base pairs of PCR product size).

3 Monaghan G, Ryan M, Seddon R, Hume R, Burchell B. Genetic variation in bilirubin UDP-glucuronosyltransferase gene promoter and Gilberts syndrome. Lancet 1996;347:57881. 4 Brown WR, Boon WH. Ethnic group diVerences in plasma bilirubin levels of full-term, healthy Singapore newborns. Pediatrics 1965;36:745-51. 5 Horiguchi T, Bauer C. Ethnic diVerences in neonatal jaundice: comparison of Japanese and Caucasian newborn infants. Am J Obstet Gynecol 1975;121:71-Linn S, Schoenbaum SC, Monson RR, Rosner B, Stubbleeld PG, Ryan KJ. Epidemiology of neonatal hyperbilirubinemia. Pediatrics 1985; 75:770-4.
Randomised controlled trial of cisapride in preterm neonates for gastric emptying time EDITOR,We read with interest the study on the randomised controlled trial of cisapride in preterm infants reported by by McClure et al.1 We have recently concluded a randomised, double blind, placebo controlled study to evaluate the eVect of cisapride on feed tolerance and gastric emptying of preterm neonates. (P S Reddy, A K Deorari, C S Bal, V K Paul, M Singh. Abstract 1705 presented to the Society of Pediatrics Research meeting, 1-4 May 1999, San Francisco, USA). After obtaining informed parental consent a total of 44 preterm neonates stratied by gestation into 2932 weeks and 3334 weeks, and randomly allocated to receive either cisapride or placebo at a dose of 0.2 mg/kg every 8 hours. The babies were enrolled once they were stable and receiving oral feeds amounting to 25 per cent of their uid requirements. The feeds were increased gradually in a stepwise manner, around 20 ml/kg/day. Gastric emptying time was measured on days 6 to 8 after enrolment using a dynamic technetium scan using 100 Ci of radioactivity. This results in only 0.2 mSV of whole body absorbed dose.2 To oVset the eVect of type of feeding on gastric emptying, the two feeds before the nuclear study were of uniformly expressed breast milk. Gastric emptying time (mean (SD) and median) in the cisapride group (58.1 (32.2 mins) 48.8 mins) was not signicantly diVerent from that of the control group (53.8 (34.6 mins) 43.4 mins). Clinically signicant gastro-oesophageal reux was seen in 50% of babies in each group. Cisapride had no eVect on either the number of episodes of feed intolerance or the length of feed intolerance. Weight gain and the day on which the total enteral feeding began were also comparable in the two groups. Our observations support the ndings of McClure et al, that cisapride does not improve gastric emptying in preterm neonates and that its use for establishing enteral feeds is not warranted.

A K DEORARI P S REDDY V K PAUL Department of Paediatrics All India Institute of Medical Sciences New Delhi 110029 India C S BAL Department of Nuclear Medicine 1 R J McClure, J H Kristensen, A Grauaug. Randomised controlled trial of cisapride in preterm infants. Arch Dis Child Fetal Neonatal Ed 1999;80:F174-F7. 2 Heyman S. Gastroesophageal reux, gastric emptying, gastrointestinal motility.In: Treves ST, ed. Pediatric Nuclear Medicine. 2nd Edn. Springer-Verlag, New York, 1994: 430-52.

Table 1

Clinical data and DNA polymorphism of all infants
Healthy infants TA-6/TA-6 TA-6/TA-2/(184) 39.6 (1.4) 5/2 9.2 (1.5) TA-7/TA-0/38.9 1/0 9.1 Infants with jaundice TA-6/TA-10/(255) 39.0 (1.2) 10/8 18.8 (2.29) TA-6/TA-0/38.0 1/0 15.7
Number of subjects Male/female Birthweight (g) (mean (SD)) Gestational weeks (mean (SD)) Breast fed/formula fed Bilirubin (mg/dl)*
47 22/(350) 39.8 (1.1) 25/22 10.0 (2.7)
*Bilirubin concentrations at 45 days old in healthy infants and the peak bilirubin in infants with jaundice.

Letters

1 Brimacombe J. Use of the laryngeal mask airway in very small neonates. Anesthesiology 1994; 81:1302. 2 Rabb MF, Minkowitz HS, Hagberg CA. Blind intubation through the laryngeal mask airway for the diYcult airway in infants. Anesthesiology 1996;84:151011. 3 Ofer R, Dworzak H. Die Kehlkopfmaske-ein wertvolles Instrument bei erschwerter kindlicher Intubation. Anaesthetist 1996;45:26870. 4 Ogawa Y, Miyasaka K, Kawano T, et al. A multicentre randomised trial of high frequency oscillatory ventilation as compared with conventional mechanical ventilation in preterm infants with respiratory failure. Early Hum Dev 1993;32:110.
Use of laryngeal masks in the resuscitation of a neonate with diYcult airway EDITOR,Neonates with mandibulofacial anomalies and respiratory distress present a challenge for neonatologists. We report a newborn boy with severe micrognathia who failed to breathe adequately immediately after birth. Tracheal intubation was unsuccessful, but he was ventilated for several hours using a laryngeal mask. Case report The boy was born to a 29 year old primigravida after 39 weeks of gestation. During the pregnancy, absence of the corpus callosum had been noted at 33 weeks of gestation, and genetic amniocentesis had indicated a microdeletion of chromosome 21. No other malformation had been veried. Caesarean section was performed because of intrauterine growth retardation and breech presentation. At birth the boy weighed 2300 g, had multiple contractures of the limbs, bilateral coloboma of the iris, severe mandibular hypoplasia with a small oral orice and a massive glossoptosis, and a systolic heart murmur. He made feeble attempts to cry, but remained cyanotic and bradycardic despite the jaw thrust manoeuvre and bagging. Oro-tracheal intubation with a 3 mm and a 2 mm tracheal tube failed twice. A lubricated number 1 laryngeal mask was easily put into the right position while holding the tongue with a Magill forceps. After sedation with fentanyl and midalozam he tolerated intermittent mechanical ventilation well for three hours while investigations were carried out. At ventilator settings of PI 25 cm H2O, PEEP 4 cm H2O, RR 60 per minute, FIO2 0.4, I:E ratio of 1:2, ow rate of 12 l/min, capillary blood gas analysis was pH 7.3, pCO2 4, pO2 95, base excess 5 mmol/l. Blood pressure and heart rate were within the normal range. Echocardiography revealed a severely hypoplastic distal aortic arch with a wide open ductus arteriosus, and cranial echocardiography conrmed the absence of the corpus callosum. In view of the poor prognosis the parents requested the withdrawal of intensive care, and at necropsy the diagnoses were conrmed, with the addition of a micro larynx and a large pharyngo-oesophageal cleft. No signs of damage to the hypophraynx were found. The use of the laryngeal mask in our patient gained us time to perform investigations to establish prognosis. In patients with isolated cranio-facial and mandibulo-facial malformations such as Goldenhar, TreacherCollins, or PierreRobin syndromes, the laryngeal mask could be used to establish a denite airway either through bronchoscopic intubation or tracheotomy. The usefulness of the laryngeal mask during anaesthesia in such cases has been well documented,13 but we believe that the mask could also be a valuable tool for neonatologists who unexpectedly face a diYcult airway.

RUDOLF TRAWGER CHRISTIAN MANN Department of Paediatrics University Hospital Innsbruck Innsbruck A-6020 Austria MANFRED MRTL Department of Anaesthesiology KARIN RIHA Department of Obstetrics
Antenatally diagnosed renal pelvis dilatation EDITOR,I read with interest the study by Jaswon et al on outcome in antenatally diagnosed renal pelvis dilatation.1 The cohort was recruited antenatally from pregnancies where renal pelvis dilatation had been diagnosed mostly on the 20 week ultrasound scan. VUR was described as being the most common clinically signicant pathology (23 out of 104 cases). Presumably all of the babies were asymptomatic. How can the authors be sure that this VUR was either clinically signicant or indeed pathological? Figure 2 shows that nine out of 23 babies had grade III or IV VUR despite a normal postnatal renal ultrasound scan. Is anything known of the natural history of these nine babies? The reference regarding the possible prevalence of VUR in infants is a literature review2 (method not stated by the author) of 14 papers published between 1916 and 1967that is, predating the era of antenatal ultrasonography. One baby went on to have pyeloplasty at 18 months of age, because of deteriorating renal function, due to PUJ obstruction, not VUR (table 1). This low rate of surgical intervention (one baby out of 139 pregnancies over 18 months) reinforces the notion that these ndings are largely benign.3] The longer term follow up, or outcome, of the other 103 babies is not statedfor example, if the postnatal renal ultrasound scan and MCU were normal, n=60/104, the infant was discharged. This suggests follow up only of babies with abnormal ndings. Information has been collected regarding the prevalence of VUR, PUJ obstruction, and renal dysplasia in the rst three months of life in this cohort. What is the evidence that these diagnoses have clinically important long term adverse outcomes? The conclusion states that babies with antenatal renal pelvis measurements of 5 mm or greater should be investigated, as they may have VUR. However, I am unclear as to what long term outcome measures will be improved by these sometimes invasive investigations. Table 1 also contains data on 25 babies, but the text refers to persisting renal pelvis dilatation in 47 babies.
R M NICHOLL Department of Paediatrics North West London Hospitals NHS Trust Northwick Park Campus Harrow HA1 3UJ 1 Jawson MS, Dibble L, Purie S, Davis J, Young J, Dave R, Morgan H. Prospective study of outcome in antenatal diagnosed renal pelvis dilatation. Arch Dis Child Fetal Neonatal Ed 1999;80:F1358. 2 Bailey RR. Vesicoureteric reux in healthy infants and children. In: Hodson J, Kincaid Smith P, eds. Reux Nephropathy. New York: Masson, 1979:5761. 3 KoV S, Campbell K. Non-operative of unilateral neonatal hydronephrosis. J Urol 1992;148: 5253.

Multicentre trial of high frequency ventilation EDITOR,Cools and OVringas recent metaanalysis of elective high frequency ventilation (HFV) in preterm infants with respiratory distress syndrome1 concludes that HFV reduces the risk of chronic lung disease (CLD) at 36 weeks of postconceptional age, but may be associated with an increased risk of severe intraventricular bleeding. Several areas of uncertainty remain, however, and they suggest that new clinical trials should be done in very preterm infants to evaluate the usefulness of elective HFV, using a high lung volume strategy, started as soon as possible after birth. We are currently running exactly such a trial in the UKthe United Kingdom Oscillation Study (UKOS). This is a multicentre trial comparing high frequency oscillatory ventilation (HFOV) with conventional ventilation in preterm infants < 29 weeks of gestation. Previous trials have included more mature babies, but we have restricted recruitment to those babies with the highest incidence of chronic lung disease and of neurological complications. We expect to recruit 1200 babies over two years, making this the largest study of its kind so far, with the greatest statistical power. To avoid selection bias, treatment allocation is by a central telephone randomisation service. Infants are given their allocated mode of ventilation within one hour of birth (up to 15 hours in previous studies) to assess the eVect of early intervention with HFOV, which has been shown to be most benecial in animal studies.2 Long term neurodevelopmental and pulmonary outcome will be evaluated up to 2 years of corrected gestational age; only two previously published studieshave done this. Keszler and Dunn in North America (personal communication) also have an ongoing study of a similar nature, and we hope that the results of these trials will provide the evidence for future ventilation policy for very preterm infants.
A JOHNSON S CALVERT N MARLOW A GREENOUGH UKOS STUDY GROUP St George s Hospital Medical School London SW17 0RE 1 Cools F, OVringa M. Meta-analysis of elective high frequency ventilation in preterm infants with respiratory distress syndrome. Arch Dis Child Fetal Neonatal Ed 1999;80:F15F20. 2 Meredith KS, de Lemos RA, Coalson JJ, et al. Role of lung injury in the pathogenesis of hyaline membrane disease in premature baboons. J Appl Phsiol 1989;66:21508. 3 Hi Fi Study Group. High frequency oscillatory ventilation compared with conventional mechanical ventilation in the treatment of respiratory failure in preterm infants. N Engl J Med 1989;320:8893.