AD-R142 927 COMPUTER-ASSISTED STATISTICALANALYSIS, III THE i/ibROSOPHILA IELANOGASTER SE.. U LETTERMAN RRMY INST OFRESEARCH PRESIDIO OF SAN FRANCISCO CA

UNLASFID N R POWERS ET AL. AUG 83 LAIR-83-45TN F/G 6/20 N

IMMMLI

Kl

1. 2y 5I:F mL. 1. 6 ~-~~x .

IIIII. im

Lwi

TECHNICAL NOTE NO. 83-45TN

COMPUTER-ASSISTED STATISTICAL 'ANALYSIS.Ill. The Drosophila rmelaogater Sex-Linked Recessiv Letha Assay

NNELSON R. POWERS, PhD, CPT MSC

Nan__ WARREN W. JEDERBERG, US, CPT MSC

TOXICOLOGY GROUPDIVISION OF RESEARCH SUPPORT

UJI

AUGUST 1983 Toxicolmg Soris 61

LETTERMAN ARMY INSTITUTE OF RESEARCHPRESIDIO OF SAN FRANCISCO, CALIFORNIA 94129

84 07 06 002

Computer-Assisted Statistical Analysis. III. The Drosophila melanogaster

Sex-Linked Recessive Lethal Assay (Toxicology Series 61)--Powers andJederberg

Reproduction of this document in whole or in part is prohibited except with the permission of theCommander, Letterman Army Institute of Research, Presidio of San Francisco. California 94129.However, the Defense Technical Information Center is authorized to reproduce the document forUnited States Government purposes.

Destroy this report when it is no longer needed. Do not return it to the originator.

Citation of trade names in this report does not constitute an official endorsement or approval of theuse of such items.

This material has been reviewed by Letterman Army Instituteof Research and there is no objection to its presentation and/or publication. The opinions or assertions contained hereinare the private views of the author(s) and are not to be con-strued as official or as reflecting the views of the Department.,t the Army or the Department of Defense. (AR 360-5)

........................................... ...... o~a................. ;)at re an date)

This document has been approved for public release and sale; its distribution is unlimited.

UNCLASSIFIEDSECURITY CLASSIFICATION OF THIS PAGE (WIfh Doe Entered)

REPORT DOCUMENTATION PAGE READ INSTRUCTIONSREP T DBEFORE COMPLETING FORM

1. REPORT NUMBER 2. GOVT ACCESSION NO. 3. RECIPIENT'S CATALOG NUMBER

LAIR Technical Note No. 83-45TN . /4. TITLE (ad Subtitle) S. TYPE OF REPORT A PERIOD COVERED

Computer-Assisted Statistical Analysis. III. FinalThe Drosophila melanogaster Sex-Linked Recessive Dec 79 - Dec 80Lethal Assay 6. PERFORMING ORG, REPORT NUMBER

. 7. AUHRs . CONTRACT OR GRANT NUMBER(*)

Nelson R. Powers, PhD, CPT MSCWarren W. Jederberg, MS, CPT MSC

" . 3. PERFORMING ORGANIZATION NAME AND ADDRESS 10. PROGRAM ELEMENT. PROJECT, TASK" ° -"AREA & WOR UNIT NUMESToxicology Group, Division of Research Support AU

Letterman Army Institute of Research 3M167701871Presidio of San Francisco, CA 94129 WUL 203

11. CONTROLLING OFFICE NAME AND ADDRESS 12. REPORT DATE

August 1983. 13. NUMBER OF PAGES

1614. MONITORING AGENCY NAME & AODRESS(II dllferent fram Controlling Office) IS. SECURITY CLASS. (of this report)

UNCLASSIFIED

ISA. DECLASSIFICATION/DOWNGRADINGSCHEDULE

I6. DISTRIBUTION STATEMENT (of this Report)

THIS DOCUMENT HAS BEEN APPROVED FOR PUBLIC RELEASE AND SALE: ITS DISTRIBUTION

IS UNLIMITED...

* N.17. DISTRIBUTION STATEMENT (of the ab tract entered In Block 20, Ii different from Report)

I0. SUPPLEMENTARY NOTES

II. KEY WORDS (Conminue an reveee ide If necoeem and Identify by block numiber)V.

Drosophila melanogaster, Sex-Linked Recessive Lethal Assay, Computer-Assisted

Statistical Analysis, Automated Data Analysis

a.24L ASSTRACT' (Mirntie si reverse aide N aeseemy ss identf by block nuiber)

This report contains computer programs which summarize the data. One program' calculates the mean mutation frequency with its variance expressed as a percent

and the other executes statistical analyses of the data obtained from theDrosophila melanogaster Sex-Linked Recessive Lethal Assay.

DD, 1403 EIONw OF 9 NOV 4s s1 OBSOLETE UNCLASSIFIED

4 07 06 0 2, SECUMTY CLASSIFICATION O THIS PAGE (When Data ... r-d

47:N.

PREFACE

This technical note is the third in a series on the utilization ofthe computer to assist mutagenicity testing at Letterman ArmyInstitute of Research. In this report we detail three functions:searching of a specific file to summarize the data; calculating themean mutation frequency with its variance expressed as percentages;and using a statistical program for analysis of the data. Using theprogram in the Drosophila melanogaster Sex-Linked Recessive LethalAssay provides an easy manner to view the data and perform statistical

analyses.

W1

N -S RIon For

~~' 0

,..

-~bDTSpecial

Powers-- 1

Computer-Assisted Statistical Analysis MII.

The Drosophila melanogaster Sex-LinkedRecessive Lethal Assay

The Drosophila melanogaster Sex-Linked Recessive Lethal Assayrequires the testing of 8000 to 10,000 X-chromosomes for each compound

and concurrent negative control in addition to 500 X-chromosomes for apositive control. A mass of data must be processed. Therefore, twocomputer programs were developed to summarize the results. By

totalling the number of lethal mutations and non-lethal mutations foreach brood of each replicate yielding the total number of lethalmutations and number of tests, this first program calculates the

percentage mean mutation frequency and its variance. The secondprogram erforms the statistical analysis using the BMDP statisticalsoftwareym package by testing the null hypothesis if the number of

lethal mutations resulting from the compound is non-significantly fromdifferent from the number resulting from the negative control.

DESCRIPTION OF THE PROGRAM

The program DRUNSUM1.FR (Appendix A) searches the data file andsummarizes the data accumulated in the program DROSTOXDATA.FR (1) fora specified subgroup of a given run number selected by the user. Thisprogram uses the subroutines RDDROSDATA.FR and STRING.FR (Appendix B

'Uand C) to read the data sequentially and accumulates the valuesnecessary for the number of lethal, non-lethal, and total testsperformed for each brood, and grand total for the specified run

*: number. The program also calculates the mean mutation frequency andits variance and displays them as percentages. The detection of threeerror conditions is provided to see (a) if more than 25 records arepresent for the specified subgroup within the run; (b) if the file

name under which the data from the DROSTOXDATA are stored is present;(c) if the specified subgroup is present.

The program DRUNSUM.FR (2) requests the following information fromthe user (Figure 1):

*The name of the file to be searched.

*The run number to be searched.*The group designator of the male code number (this specifies

if the male was from the control, positive, or test compound).

The second program is part of the BfDP statistical softwareTM (3).Instructions for its use are given in this statistical package. Thispackage analyzes the data in a 2 x 2 table and two statistics are

used, the Yates corrected chi-squares and Fisher's Exact test. TheYates correction is intended to improve the approximation to thechi-square distribution. The Fisher's Exact test is computed when the

'T I T I _ . , . . " b. -.. -.. - . --.

4Powers--2

minimum value is less than 20. Both 1-tail and ?-tail probabilitiesare tested. However the 2-tail probability for Fisher's Exact test

should be used.

The program requests the following information !turi the user:

*Enter the number of lethal, lethal mutations and non-lethalmutations resulting from the negative control.

*Enter the number of lethal mutations and the number of

non-lethal mutations resulting from the concurrent test compound.

The computer will then type the values and prubztilities assciated.4ith the Yates corrected chi-square and the Fisher's Exact test.

kOMMENT

By utilizing these programs and the subroutines presented in thisrtport, the raw data may be summarized and statistical analysis

per formed.

CONCLUSION

None

!: COM1ENDATION

None

Powers-- I

REFERENCES

1. Jederberg WW, Powers NR. Computer-assisted data storage and

retrieval in mutagenicity testing II. The Drosophila melanogaster

sex-linked recessive lethal assay. Technical Note No. 82-42-TN.Presidio of San Francisco, CA: Letterman Army Institute ofResearch, 1982.

2. LAIR SOP-OP-STX-63, Execution of FORTRAN V Program DRUN: 14 June1982.

3. Brown B14. Frequency tables. In: Dixson WJ, Brown MB, Engleman L,Frane JW, Hill MA, Jennrich RI, Toporek JD, eds. BMDP statisticalsoftware, Berkeley: University of California Press, 1981:1943-161.

Powers--)

SAMPLE RUN OF DRUN

) DRUNThe computer will type:

NAME OF FILE TO BE SEARCHED: DROSDATA

The computer will type:****DROSOPHILA RECORD SEARCH AND SUMMARY****PLEASE ENTER RUN NUMBER TO BE SUMMARIZED: 24

The computer will type:ENTER GROUP TO BE SEARCHED FOR:(TI, T2, P1 or C1 ETC. ALLOWED): C1

The computer will type:

SUMMARY TOTALSRUN NUMBER: 24GROUP TYPE: C1

BROOD DATA

LETHALS: 0 1 0 1

NONLETHALS: 596 524 547 547

GRAND TOTAL LETHALS: 2GRAND TOTAL NONLETHALS: 2214

TOTAL RECORDS PROCESSED: 25

TOTAL TESTS REPRESENTED: 2216* lEAN MUTATION RATE: .08000%, WITH VAR. .07667%

The computer will then display I)", the user may then return to

the program for the next file to be searched or log off.

i-,

Figure 1. Sample Run of DRUN(Underlined entries are sample user inputs)

- " '#;,I :

- * -*,' -,:""''',,,., 'e',/""',j;,''',e'',; ' - ,." ., -•• - . . .- .

Powers-- /

Page

APPENDIX A. Listing for DRUNSUM.FR ............................... 9

APPENDIX B. Listing for subroutine RDDROSDATA.FR ................ 13

APPENDIX C. Listing for subroutine STRING.FR. .................... 14

Powers--9

APPENDIX A

C *** PROGRAM NAME: DRUNSUI.FRC *** WRITTEN BY: WARREN W. JEDERBERGC *** PURPOSE: TO SUICARIZE THE DATA FROM SPECIFIEDC *.' RUNS IN DROSDATA.

C *** THE SUBROUTINE RDDROSDATA IS USED TO READ EACH RECORDC *** VARIABLE DEFINITIONSC N =NUMBER OF RECORDSC FBL =FIRST BROOD LETHALSC FBNL=FIRST BROOD NONLETHALSC STL =SECOND BROOD LETHALSC STNL=SECOND BROOD IONLETHALSO TTL =THIRD BROOD LETHALSC TTNL=THIRD BROOD NONLETHALSC FTL =FOURTH BROOD LETHALSC FTNL=FOURTH BROOD NONLETHALS

c

C GTT = GRAND TOTAL TESTSc GTL =GRAND TOTAL LETHALSC GTNL=GRAND TOTAL NONILETHALSC T4R =TOTAL MUTATION RATEC V1I4R=ACCUMULATED MUTATION RATES SQUAREDC NNR =DESIGNATED RUN NUMBER (USER INPUT AT RUN TIME)C NS =SUBGROUP DESIGNATOR (USER INPUT AT RUN TIME)C RMR -MEAN MUTATION RATECC VARIABLES MENTIONED IN THE COTION STATEMENT ARE DEFINED INC THE FORTRAN V PROGRAM DROSTOXDATA.

INTEGER NR,MNUM,CNAM,BFF,BFL,BFNL, BSFBSL,BSNL,BTF,BTL,BTNL,IBLF,BLL,BLNL,TF,SL,ML,TL,TNL,TT,SFNAM.

DIMENSION NS(2),SFNAMI(20)REAL MXR

CO MON/DREC/NR,r.iNUM(6),CNAM(6),BFF,BFL,BFNL,BSF,BSL,BSNIL,BTF,BTL,IBTNL,BLF,BLL,BLNL,TF,SL,ML.TL,TNL,TT, IR

-. . . . sis. P.e... X

Powers--1O

APPENDIX A(continued)

.. .OGICAL EOF

C ** SET ACCUeqJLATORS TO 0.00N-0FBL-OFBNL=OSTL-0STNL=OTTL-OTTNL=OFTL0FTNL=OGTT-OGTL-OGTNL=0TMR=OV I MR -0

C *** INITIALIZE EOFEOFS. FALSE.

C * GET FILE NAME TO CHECK FOR THE DATAWRITE (10,220)

220 FOR.4AT ('"NAME OF FILE TO BE SEARCHED: ,-)READ (11,210) SFNAM

210 FORMAT (2CA2)CALL STRING (SFNAf1,20)OPEN 1 ,SF; NAX,ATT-"SIB',ERR-1O0READ (1,5)

5 FORI-IAT (/)GO TO 120

100 WRITE (10,6) SFNAM6 FORMAT (//,"*** ERROR **** ERROR **** ERROR ****",/,

14X,"FILE ",20A2," NOT FOUND . . ."

a- STOP120 CONTINUE

C *** PRINT HEADER AND GET RUN NUMBERWRITE (10,1)

1 FORMAT (/,1OX,"**** DROSOPHILA RECORD SEARCH AND SU,''.ARY ***, ,

110X,"PLEASE ENTER RUN NUMBER TO BE SUMMARLED: ,,')

ACCEPT TNR

V , . . ,.r ,.t..,,..;:-;..,-.. ,,,:.,,f; . .a , .. -,''~' '

Powers-I

APPEN~DIX A(continued)

C GET GROUP DESIGNATOR99 CONTINUE

WRITE (10,3)3 FOR?4AT (/,1OX,"ENTER GROUP DESIGIATOR TO BE SEARCHED FOR ,/11OX,"(Tl,T2,Pl,OIR Cl ETC. ALLOWED):",Z)

V READ (11 ,4)(INS(I),I=1 ,2)4 FORMAT (WA)

IF (NS(1 ).NE.'c ".AND.NS(1 ).NE."P ".AND.NS(1 ).NE.*T ")GO TO 99C ** READ DATA RECORDS

140 CALL RDDROSDATA (EOF)IF (EOF) GO TO 225IF (NR.NE.NNR) GO TO 140IF (NS(1).NE.flNUM(1)) GO TO 140IF (NS(2).NE.r.tNUI4(2)) GO TO 140N=N+1IF (N.GT.25) GO TO 500FBL=FBL+BFLFBNL=FBNL+BFNL

p. STL=STL+BSLSTNL=STNL+BSNLTTL=TTIJ+BTLTTNL-~T'NL+BTNLFTL=FTL+BLLFTNL-FTNL+BLNLGTL-GTL*TLGTNL=GTNL+TNLGTT=GTT*TT

V1I"wVlMFs(mPt**2)k GO TO 140

C ** IF N>25 ERROR500 CONTINUE

WRITE (10,7)7 FORMAT (/,"*** ERROR ERROR ERROR

14X,"MORE THAN 25 RECORDS FOUND...STOP

iPowers-l 2

APPE:IDIX A(continued)

225 CONTINUEC ** IF NO SPECIFIED RECORDS FOUND

IF(N.NE.O.O) GO TO 300'WRI;E (10,9) NNR,NS,SFNAM

9 FORM~AT (/,"** ERROR *** ERROR ERROR

24X,"NOT FOUND IN ,1X,20A2,"..STOP

300 CO NTI NUEC CALCULATE 10HUATION RATE AND VARIANCE

R!4Mw (THRu/N)

C WRITE OUT TOTALSWRITE (10,S) IlR, 'S, ,FBL, STL,T1TL, FTL. FB.L, STNTLTTN L, FTNL,

j lGTL,GTIlL,I;,GTT,RTMR,VN.R

8 FORMAT (20X," SUK1WARY TOTALS:%/.

210X,"GROUP TYPE: ",2X,2Al$ I,320X,"BROOD DATA:",!,

57X,"NONLETHALS: " 4(5X,I3) ,/1,610X,"GRAND TOTAL LETHALS:",5X,I5,/,77X, "GRAND TOTAL NO0ULETHALS:" ,5X, 15//,810X,'TOTAL RECORDS PROCESSED:",5X,I2,//,91 OX, "TOTAL TESTS REPRESENTED:" ,5X, 15,/,llOX,--NrEAN I'FTATION RATE: ",IX,F1O0.5,"% WITH VAR.", lX,F1O.5,"",/)

CLOSE IEND

iowers-li

APPENDIX B

COMPILER NOSTACKC PROGRAM NAIKE: RDDROSDATA.FR

C ~'WRITTEN BY: WARREN JEDERBERGC PURPOSE: TO READ DATA FOR tMAKIANG REPORTSC FOR THE SLRL-DROSOPHILA ASSAY

SUBROUTINE RDDROSDATA (EOF)COMMON/DREC/NR, r-NUM(6) ,CNAM(6), BFF, BFL, BF'.L, BSF, BSL, BSN"L,I BTF,BTL,BTNL,BLF,BLL,BLNL,TF,SL,ML,TL,TNL,TT,MIR

INTEGER NR r"U CN]LI, BFF, BFLA,BFNL, BSF, BSL, BSIL, BTF,-BT T. BTNL,5% BLF,BLL,BLNIL,T.F,SL,I--L,TL,TNL,TT

REAL MR

LOGICAL EOF

READ 0 , 100, END= 120) NR, "2 UM, C!AM, BFF, BFL, BFNL, BSF, BeL. BS:;L,IBTF,BTL,BTIIL,BLF,BLL,BLNL,TF,SL,McL,TL ,Tl,'L,TTA,K

100 FORMAT(I4,lX,6Al,lX,6A1,4(1X,3I2),IX,I2,1X,31&2,12(11,13) ,lX,F6.2)RETURN

120 EOP w.TRUE.

END

U I Aq * 3 0! 0

APPENDIX C

C TO CONVERT FILE NAMES TO USE IN OPEN sTAT.E,.:EirrsSUBROUTINE STRING (LINE,LLEN)

xz COr.PILER STATICDIMENSION LINE (LL:zu)DO 100 1-1,LLENIF (LINE(I).EQ." 'a) LIIIE(I)=O

100 CONIT INUERETURNEND

-U ~ ~ ~ -~ ~ .------------ c . - . - - - - . -

A .' 4i~ 'I . *

Pt ~ -4.., 'I.-

9 4j

t '~ ---- ~~

I .- '%

N.

I .~ ,'9 _ K Lt . N4" 14. r...41.jV~~ ~ S

.-- 1 ~ *

~-~- .. V'ta. .j '.*~ t 4v~t K--

* ~ ~r'~ * ~MW .~ ~..rr< <w' - -~

-- M 44*B -C

* . ;A--4 ~.(-J~ j, . .9..

* -. '1-*i~' , 44 9.%~ '~ t t . . . .9 A.

A' . 4*~* ~

IS( r

.-. ,v y 14Vr4 IM~

~,, ~

* . i-- It IF

-. 4, . .rtu Pit) i -**

- *1 -- - ,- 9

t. '44 J~- * . -~

I I 1r2vf H ~ nAj.,4 -1 i I

"if ~ ~....

C-.' A.

9-? ~ It ~,* r

$~A' 0 '~ clJo - -- .9 ',;~~~;~-4- ~'

- ~ 'k~ha-'h-i.

#I~A ' /

a*J V. I. *

*1-

7 24kt . Kj~~' ., *

~~ ~ jannfl~j t>Kf+< ~ j*qt'. 'p -

9%- - . -

- *... fib ...

- *-.±1 -

.9 I

'<

s~i ~%k. ~ Fvw4~tt~ F'~?j WI- &

%~* 1. 4' 7-f;at

-(S.

Ste * * - **t. -- '. * ta stt~ t.taw..ta*.Z<