Rapid Analysis of Electronically Monitored Homopteran Feeding
Contents
1. 1992 Windaq and Windaq EX users guide software release level 1 31 manual revision C Dataq Instru ments Akron OH 1993 DI 200 PGH PGL software configurable analog and digital I O board user s manual manual revision C Dataq Instruments Akron OH 1994 Windaq 200 user s manual manual revision C Dataq Instruments Akron OH Febvay G Y Bonnin R Bournocille S Delrot and J L Bonnemain 1988 Resistance of different lucerne cul tivars to the pea aphid Acyrthosiphon pisum influence of phloem composition on aphid fecundity Entomol Exp Appl 48 127 134 Febvay G Y Rahb and M van Helden 1996 Mac Stylet software to analyse electrical penetration graph data on the Macintosh Entomol Exp Appl 80 105 108 Gruenhagen N M and E A Backus 1996 An inexpen sive analog to digital system for computer assisted measurement of electronic monitoring waveforms J Agric Entomol 13 73 78 Hardie J M Holyoak N J Taylor and D C Griffiths 1992 The combination of electronic monitoring and video assisted observations of plant penetration by aphids and behavioural effects of polygodial Entomol Exp Appl 62 233 239 Kimsey R B 1994 Measuring and recording arthropod blood feeding pp 122 128 In M M Ellsbury E A Backus and D L Ullman eds History development and application of AC electronic insect feeding mon itors Thomas Say Publication in Entomology Proceed ings Entom2. Recording B is of an adult female of E kraemeri leafhopper feeding on P vulgaris common bean ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol 91 no 1 the system developed by Gruenhagen and Backus 1996 the complete data file must be converted to ASCII format after which the user must add a time column to the variable voltage column In addition the user must keep track of probe numbers manu ally All of the data acquisition programs described above were written by the users Commercially available data acquisition software is likely to be more sophisticated than these programs but are relatively expensive gt 2 500 However the high initial costs of professional data acquisition software is compensated for by the considerable time savings of researchers not having to develop their own soft ware Most data acquisition software provides real time display of waveforms while the user is pre sented with various tools to explore the signal after recording For instance in WinDaq EX it is possi ble to zoom in on sections of the recorded signal to examine waveform transitions in more detail In addition to cost the lack of specific tools for the analysis of EMIF data has been another obstacle for the use of professional data acquisition software Use of the EMIF workbook removes this obstacle and has the following advantages over currently available systems Waveform Code Flexibility Unlike STYLET the EMIF
3. calcula tion of the total number of occurrences the total and average duration and the percentage of the total recording for each waveform Fig 4B Also the start of each subsequent probe is determined as the transition between baseline and any other wave form Fig 4D and the total number of probes the total and average probe duration and the percent age of the whole recording spent probing are cal culated Fig 4E As was the case in worksheet III x or signs will appear in the column left of the waveform codes Fig 4H to indicate coding and duplication errors respectively In the example given in Fig 4 no errors were detected After eliminating any coding or duplication errors in worksheet III the internal checks percent to tal count and probe at the top of worksheet IV should indicate OK and CE number of coding errors and DE number of duplication errors should be 0 Fig 4G If this is not the case there may be an error in the cumulative time column e g not all numbers increase successively going down the time column which may require editing of the original EMIF data file before import into worksheet Il Worksheet V Summary Apart from the experi ment specific variables Fig 5A worksheet V lists the waveform parameters Fig 5B and probing parameters Fig 5C calculated in worksheet IV The user can print out the summary by pressing a button on t
4. with a custom set of variables and waveform codes under a unique filename Worksheet II Data EMIF data files are im ported into the EMIF workbook by pressing 1 of 2 buttons positioned on worksheet III Fig 3A The button labeled WNF Format lets the user import WinDaq Notepad Files and the button labeled EMIF data File permits the user to load any data file conforming to the aforementioned 2 column January 1998 IV Analysis Experiment File E435 T60401F 07 25 97 WAvG VAN GIESSEN AND JACKSON ELECTRONICALLY MONITORED FEEDING BEHAVIOR Date Experimenter eine itl meg a Morph 149 percent total count pone 7a a al vir OK OK OK eT percentage gt 2 5 26 9 2572 31 1 i sees SE a a a a E E TE a NS a E Ne ON 8 ON Fig 4 982 4 1964 7 percent gt B avg dur gt total dur gt total no gt DOT NANNA BP ke RWW WY DY HH A 0O Example of worksheet IV of the Excel workbook for the analysis of electronically monitored insect feeding behavior In this worksheet data files indicated in gray A are analyzed automatically after import Parameters such as total number total duration average duration and percentage total duration of a waveform total duration of recording are calculated for each waveform B through automatic sorting of the waveform durations from the datafile into the respective worksheet columns assigned to each waveform C Also the start of each suc
5. BEHAVIOR Rapid Analysis of Electronically Monitored Homopteran Feeding Behavior WILANT A VAN GIESSEN anb D MICHAEL JACKSON U S Vegetable Laboratory USDA ARS 2875 Savannah Highway Charleston SC 29414 5334 Ann Entomol Soc Am 91 1 145 154 1998 ABSTRACT Historically a bottleneck in the analysis of electronically monitored insect feeding behavior EMIF of homopterans has been the conversion of raw waveform data into a format that is easy to analyze Here we present an efficient method for the analysis of EMIF recordings using Microsoft Excel a popular spreadsheet program Recordings were made with WinDaq data acquisition software from Dataq Instruments Waveform transitions were saved to file using WinDaq s annotation utility Subsequent data analysis of the WinDaq data file was performed using a custom designed Microsoft Excel EMIF workbook After experiment specific variables are entered and the data file is imported by the user a complete analysis is performed automatically The Excel workbook summarizes the results and generates an output file for further statistical analysis Any EMIF data file containing waveform codes and cumulative time i e time from the beginning of the recording in seconds can be analyzed The new method described in this article was used to analyze EMIF recordings of 2 homopteran species Aphis gossypii Glover Aphididae and Empoasca kraemeri Ross amp Moore Cicadellidae Identification an
6. ardware DI 200 plug in board Dataq Instru ments 1993 and DI 205 external signal I O panel 16 channels 16 bit resolution 25 Hz sampling rate 82 900 Hz max Dataq Instruments Akron OH All recordings could be viewed simultaneously and in real time Post acquisition waveform identifica ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol 91 no 1 tion was performed using WinDaq EX playback and analysis software Waveform Identification In WinDaq EX Dataq Instruments 1992 Fig 1A the data cursor Fig 1B can be moved along the recorded signal using the mouse or the left and right arrow keys With the cursor at the start of the recording the keyboard command Al n will bring up the Notepad File win dow that will prompt the user for a filename use extension WNF for WinDaq Notepad File Use of the Notepad File function is simple and is described in detail in the WinDaq manual Its use will be briefly summarized here After bringing up the Notepad File Function by hitting Alt n at the be ginning of the EMIF data file the user is prompted to enter the code for the waveform recorded at time 0 which is typically the baseline The choice of codes for the different waveforms is entirely up to the user However waveform codes must be 1 5 characters long and should all be of identical length Next the data cursor is moved along the signal to the beginning of the next waveform and after pressing Alt n the user enters t
7. cessive probe is automatically determined D and a number of probe parameters are calculated E To the right of the experiment specific variables listed at the top of the worksheet F a number of checks G provide the user with an easy way to check for errors CE coding errors DE duplication errors the labels percent total count and probe refer to 4 internal checks In addition x or signs will appear in the column headed E next to the waveform codes H to indicate coding and duplication errors respectively similar to worksheet III see Fig 3 In the example in this figure no such errors were detected formats Import of both file types is performed using Microsoft Excel s Text Import Wizard a Wizard is an interactive feature in modern software guiding the user through a multistep procedure While using the Text Import Wizard to import WNF files the user should designate the comma as delimiter For import of any other EMIF data files the comma tab semicolon space or any user defined delimiter can be used The user also can enter data directly into worksheet III or copy and paste data from another Excel file or from another application In these cases the user must add the word end to the bottom of the column with the waveform codes during automated data import as described above end is added automatically After data import coding and duplicatio
8. crosoft Windows 3 1 or Microsoft Windows 95 NT The EMIF workbook can also be used with Mi crosoft Excel for Macintosh M Serrano personal communication The file size of the EMIF work book is 2 02 MB At least 16 MB of RAM is required for the workbook to run properly EMIF Analysis The method described in this article was used to analyze EMIF recordings of 2 homopteran species Recording A was made of an apterous virginoparous cotton aphid Aphis gossypii Glover Homoptera Aphididae feeding on Mex ican lime Citrus aurantifolia Swingle provided by Ray K Yokomi Horticultural Research Laboratory USDA ARS Orlando FL EMIF recording B not shown was made of an adult female leafhopper Empoasca kraemeri Ross amp Moore Homoptera Ci cadellidae feeding on Phaseolus vulgaris L com mon bean cultivar BAT41 provided by Miguel S Serrano Department of Entomology University of Missouri Columbia MO To determine the speed with which these 2 recordings could be analyzed we timed the various analysis steps Results and Discussion Previously developed software for acquisition and analysis of EMIF data has certain disadvantages The software written by Noyes et al 1990 was found to be incompatible with other computers J C Reese personal communication and Stylet Tjallingii and Mayoral 1992 MacStylet Febvay et al 1996 and the system developed by Gruenhagen and Backus 1996 do not provide real time on scre
9. ctronic monitor for the study of aphids as vectors of barley yellow dwarf virus pp 106 121 In M M Ellsbury E A Backus and D L Ullman eds History development ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol 91 no 1 and application of AC electronic insect feeding mon itors Thomas Say Publications in Entomology Pro ceedings Entomological Society of America Lanham MD Tjallingii W F 1978 Electronic recording of penetra tion behaviour by aphids Entomol Exp Appl 24 721 730 1985 Electrical nature of recorded signals during stylet penetration by aphids Entomol Exp Appl 38 177 186 1988 Electrical recording of stylet penetration activi ties pp 95 108 In A K Minks and P Harrewijn eds World crop pests Aphids their biology natural ene mies and control vol 2A Elsevier Amsterdam 1998 Comparison of AC and DC systems for electronic monitoring of stylet penetration activities by ho mopterans In G P Walker and E A Backus eds Principles and applications of electronic monitoring and other techniques in the study of homopteran feed ing behavior Thomas Say Publications in Entomology Proceedings Entomological Society of America Lan ham MD in press Tjallingii W F and A Mayoral 1992 Criteria for host plant acceptance by aphids pp 280 282 In Proceed ings of the 8th International Symposium on Insect Plant Relationships Wageningen 9 13 March 1992 Kluwer Acad
10. d file export took an average of 5 s per waveform occurrence the total time required to identify all waveforms and export them to file will depend on the number of occurrences of each waveform in the EMIF recording quality of the recording waveform variability level of user experience and computer speed The subsequent analysis of each recording with the EMIF workbook took between 1 and 2 min Use of the EMIF workbook has the following advantages 1 it can be used to analyze EMIF recordings of any homopteran species 2 it is compatible with any data acquisition software capable of exporting data files containing waveform codes and cumulative time as well as with data files with a similar format 3 it can be used in conjunction with both AC and DC EMIF systems and 4 it is fast flexible and user friendly KEY WORDS electronic monitoring electrical penetration graph data analysis waveform spreadsheet ELECTRONIC MONITORING OF insect feeding EMIF has been used successfully since 1964 to character ize probing behavior of gt 50 homopteran species McLean and Kinsey 1964 1965 1967 for a recent review see Backus 1994 This technique has been used in studies of plant resistance to homopteran pests Febvay et al 1988 Ullman et al 1988 Montl lor and Tjallingii 1989 Calderon and Backus 1992 Webster et al 1993 Calatayud et al 1994 Cole 1994 Reese et al 1994 Wayadande 1994 Caillaud et al 1995 acquisition and trans
11. e right of this worksheet the user defines the 10 possible codes assigned to the various waveforms C and the number of characters 1 5 used for waveform codes D Depending on the speed of the computer the user can toggle between automatic and manual recalculation by pressing the button positioned next to in this case manual E Any remarks pertaining to the recording can be entered at the bottom of this worksheet F To accommodate use of the EMIF work book with monitors with different resolutions the user can select from 5 different zoom factors 80 100 120 140 and 160 G In addition each worksheet has a Sheet For ward gt gt and a Sheet Backward lt button H to move forward and backward through the EMIF workbook Even though these buttons are present on all 7 worksheets they are shown only in this figure Fig 2 Up to 10 waveform codes can be entered in no particular order The user must specify the num ber of characters 1 5 used for waveform codes below this list Fig 2D As noted previously all waveform codes must have the same number of characters Any remarks pertaining to the recording can be entered at the bottom of this worksheet Fig QF Each time the user enters information Microsoft Excel automatically recalculates all the formulas contained in all the worksheets in the workbook Automatic calculation will substantially slow down analysis on older computers Therefo
12. ee Fig 5b and are presented graphically in worksheet VII see Fig 6 ill Data see Fig 3 IV Analysis see Fig 4 V Summary see Fig 5a VI Statistics EMIF Data Fil ER E see Fig 5b VII Graphs see Fig 6 Worksheet I Instructions The procedure the user needs to follow to analyze an EMIF data file successfully is presented in worksheet I Worksheet II Specifics Experiment specific vari ables Fig 2A and 3 custom variables Fig 2B can be entered in worksheet II The names custom1 custom2 and custom3 can be changed to what ever the user desires On the right side of worksheet II the codes that the user wants to assign to each waveform described above under waveform iden tification are entered Fig 2C The Ist code Pre entered must be for baseline z in the example in 148 ll Specifics e code length Fig 2 Example of worksheet II of the Excel workbook for the analysis of electronically monitored insect feeding behavior Only the shaded areas contain information en tered by the user The information in this figure is merely a specific example On the left side of this worksheet the user enters experiment specific variables such as experi ment code filename date etc A and 3 custom variables B in this example custom variable 1 has been assigned the name morph On th
13. emic Dordrecht Ullman D E C O Qualset and D L McLean 1988 Feeding responses of Rhopalosiphum padi Ho moptera Aphididae to barley yellow dwarf virus re sistant and susceptible barley varieties Environ En tomol 17 988 991 Walker G P 1998 A beginner s guide to electronic mon itoring of homopteran probing behavior In G P Walker and E A Backus eds Principles and appli cations of electronic monitoring and other techniques in the study of homopteran feeding behavior Thomas Say Publications in Entomology Proceedings Ento mological Society of America Lanham MD in press Wayadande A C 1994 Electronic monitoring of leaf hoppers and planthoppers feeding behavior and ap plication in host plant resistance studies pp 86 105 In M M Ellsbury E A Backus and D L Ullman eds History development and application of AC elec tronic insect feeding monitors Thomas Say Publica tions in Entomology Proceedings Entomological So ciety of America Lanham MD Webster J A D R Porter C A Baker and D W Mornhinweg 1993 Resistance to Russian wheat aphid Homoptera Aphididae in barley effects on aphid feeding J Econ Entomol 86 1603 1608 Received for publication 10 June 1997 accepted 3 October 1997
14. en display during data acquisition However during use of either program the signal being re corded can be viewed on a chart recorder or oscil loscope Because recordings can last many hours and several insects may be monitored simulta neously being able to view all recordings in real time is very convenient Problems can be spotted during recording and possibly corrected Also in 152 EMIF Recording A experiment PGR PROBING file TEST145 date 3 31 95 experimenter RKY species Agoss treatment Mex_Lime insectno 1 location A3 morph alate custom 2 custom 3 Total Number Total Probing Time Average Duration Percent of Time 28 probes 10611 0 sec 379 0 sec 74 8 remarks Control Recording Waveform Dur sec Avg Dur sec baseline 3578 0 127 8 1 i 765 0 27 3 9846 0 579 2 0 0 0 0 OOnN DOA WN gt 100 0 14189 0 EMIF Recording B experiment control PROBING file 0210B44 date 2 10 95 experimenter MSS species Ekrae treatment BAT41 insect no 25 location sex female custom 2 custom 3 Total Number Total Probing Time Average Duration Percent of Time 157 probes 3518 4 sec 22 4 sec 35 2 Waveform baseline 1 OOnN DAOARWN o Fig 7 Analysis of 2 recordings of electronically mon itored insect feeding behavior Recording A is of an adult apterous virginoparous A gossypii cotton aphid feeding on C arantifolii Mexican lime
15. h a pop ular spreadsheet program EMIF recordings were made using Windaq Dataq Instruments Akron OH Waveform transitions were identified and ex ported to file using the Notepad File utility of WinDaq EX Dataq Instruments This data file or any data file containing waveform codes and cumu lative time is then imported into a custom designed Microsoft Excel EMIF workbook Excel for Win dows version 5 0 or 7 0 or Excel 97 Microsoft Redmond WA where it is automatically analyzed The EMIF analysis system presented in this article has the following advantages over existing systems 1 it takes advantage of the advanced capabilities offered by professional data acquisition software e g real time display of waveforms during record ing while Microsoft Excel is the most widely used spreadsheet software 2 the system can be used in conjunction with both AC and DC EMIF systems for use with any homopteran species and 3 the Excel EMIF workbook is fast flexible and user friendly Materials and Methods EMIF Data Acquisition The EMIF recordings that were used to develop and test the method described in this paper were made using 1 channel or 4 channel versions of an AC electronic insect feeding monitor Missouri monitor produced by the Electronic Instrument Laboratory University of Missouri Columbia Backus and Bennett 1992 Recordings were made with data acquisition soft ware WinDaq 200 Dataq Instruments 1994 and h
16. he code assigned to this next waveform This procedure is repeated for all sub sequent transitions between adjacent waveforms The format of a Windaq Notepad File WNF is illustrated in Fig 1C For each waveform code en tered by the user the WNF file automatically adds the time at which the waveform was determined to begin and the voltage levels Fig 1D in italics in all channels recorded The latter are not used in the analysis and are discarded by the Excel EMIF work book during data import Data File Analysis WNF files are analyzed using a custom designed Microsoft Excel workbook Mi crosoft Excel 5 0 Microsoft Excel for Windows 95 version 7 0 or Microsoft Excel 97 Microsoft 1993 1995 1995 1996 Fig 1H However data files from other data acquisition software or manu ally measured or compiled waveform data can also be analyzed as long as they conform to a simple 2 column format column 1 contains waveform codes 1 5 characters column 2 contains the time that each waveform begins measured in seconds from the beginning of the recording cumulative time Fig 1F or the duration of each individual waveform occurrence Fig 1G Which of these 2 data file formats is imported is automatically deter mined by the Excel EMIF workbook during import The EMIF workbook consists of a title page fol lowed by 7 so called worksheets Fig 1H To ac commodate use of the EMIF workbook with mon itors with different resolution
17. he worksheet Fig 5D Worksheet VI Statistics The information pro vided by the user in worksheet II and the results from the analysis in worksheet IV are automatically January 1998 VII Graphs Distribution S z w s 2 pa C 8 Zhs x Waveform Experiment File E435 Date Experimenter T60401F 07 25 97 WAvG Species Ap vf 4 D VAN GIESSEN AND JACKSON ELECTRONICALLY MONITORED FEEDING BEHAVIOR 151 Average Duration Duration s Z h s x Waveform Treatment Insectno Location Morph al vir Fig 6 Example of worksheet VII of the Excel workbook for the analysis of electronically monitored insect feeding behavior The distribution in time and the average duration of the different waveforms are presented in a bar graph with the experiment specific variables listed underneath combined into an output file in worksheet VI Its format facilitates appending it to similar files for further statistical analysis SAS Institute 1989 Worksheet VII Graphs The percentage distri bution of the various waveforms is presented in a bar graph in worksheet VII Fig 6 A 2nd bar graph shows the average duration of each waveform The graph can be printed out by pressing a button on the worksheet System Requirements WinDaq software or its older DOS version CODAS Dataq Instruments 1990 Microsoft Excel 5 0 Microsoft Excel for Win dows 95 and Microsoft Excel 97 run under Mi
18. ing the transition to a Windaq Notepad File WNF C Along with each waveform code entered by the user a WNF file automatically stores the time at which the corresponding waveforms were determined to begin 0 00E 00 3 46E 01 6 51E 01 and others Following the times the WNF file also automatically stores the voltage levels in italics D but these are not used in the analysis and are discarded during data import E Data import is not limited to WNF files Any EMIF file that adheres to a simple 2 column format can be imported F G The Ist column must contain the waveform codes and the 2nd column must contain either the start of each waveform from the beginning of the recording F or the duration of each individual waveform G Data files are imported E into worksheet III of the EMIF Workbook H This Microsoft Excel EMIF workbook consists of 7 individual worksheets The arrows I indicate the flow of information among the worksheets Worksheet I explains how the workbook is organized how data files are analyzed and provides the user with a guided tour through the entire workbook Experiment specific variables are entered into worksheet II see Fig 2 and EMIF data files are imported into worksheet III see Fig 3 In worksheet IV information from worksheets II and III is used to analyze the data file see Fig 4 Results are summarized in worksheet V see Fig 5a are formatted for further statistical analysis in worksheet VI s
19. ior in distinguishing the sometimes subtle waveform transitions Studying various aspects of insect plant relation ships through the use of EMIF usually involves many hours of recordings Clearly combining the strengths of WinDaq and Excel now substantially increases the speed with which these recordings can be analyzed The Microsoft Excel EMIF workbook is currently being used in the Department of Ento mology University of Missouri Columbia E A Backus personal communication and at the De partment of Entomology University of California Riverside G P Walker personal communication A copy of the EMIF workbook can be obtained free of charge from W A v G However we will be able to offer only limited technical support Acknowledgments We are grateful to Miguel Serrano and Ray Yokomi for the use of their EMIF recordings We also thank Elaine Backus Jennifer Grob Rufus Isaacs Layna Reed Miguel Serrano Freddy Tjallingii Greg Walker Sarah Alice Wyndham and Ray Yokomi for their helpful comments on the manuscript This is Technical Contribution No 4217 South Carolina Agricultural Experiment Station Clemson University Clemson SC References Cited Backus E A 1994 History development and applica tions of the electronic insect feeding monitor pp 1 51 In M M Ellsbury E A Backus and D L Ullman eds History development and application of AC electronic insect feeding monitors Thomas Say Pub licati
20. lt waveform Fig 5 Examples of worksheets V and VI of the Excel workbook for the analysis of electronically monitored insect feeding behavior Worksheet V a lists apart from the experiment specific variables A a number of parameters for each waveform B and a number of probe parameters C The user can print out the summary by pressing button D Worksheet VI b presents the data in a format suitable for appending it to similar files for further statistical analysis entered by the user in worksheet II Specifics and the codes encountered in the imported data file Codes not previously specified in worksheet II are marked with an x in this worksheet Fig 3B and in worksheet IV Fig 4H Such discrepancies could be caused either by an oversight when listing codes in worksheet II or a typographical error in the wave form column of the imported data file Similarly a waveform code listed twice in a row duplication error will be marked by a plus sign Figs 3B and 4H Pressing the button labeled Show Errors Fig 3C will reduce the visible part of the data file to just the detected errors Fig 3D facilitating correction of long data files Pressing the Show All button will again show the complete data file Worksheet IV Analysis In worksheet IV the du ration of each individual waveform occurrence is calculated and copied to the column assigned to that waveform code Fig 4C This facilitates
21. mission of plant viruses by vectors Blua and Perring 1992 Powell et al 1992 Shuckle and Quiroz 1994 arthropod blood feeding behavior reviewed by Kimsey 1994 and the ef fects of various control measures on insect feeding This article reports the results of research only Mention of a proprietary product does not constitute an endorsement or a rec ommendation by USDA for its use 1 Department of Plant Pathology and Physiology Clemson Uni versity 120 Long Hall Box 340377 Clemson SC 29634 0377 2 Department of Entomology Wageningen Agricultural Univer sity P O Box 8031 NL 6700 EH Wageningen The Netherlands behavior Saxena and Khan 1985 Hardie et al 1992 Powell et al 1993 The EMIF system originally designed by McLean and Kinsey used AC current Later Tjallingii 1978 1988 1998 developed an EMIF system using DC current and coined the term electrical penetration graph EPG for the output of an EMIF system The output from insect feeding monitors is a variable voltage signal over time This variability is caused by changes in electrical resistance across the feeding insect s stylets for AC systems or by changes in resistance in addition to biologically generated volt age sources e g membrane potentials for DC systems Tjallingii 1998 Walker 1998 These vari ations in voltage signal occur in discrete sections of uniform patterns called waveforms and are corre lated with distinct probing and feedi
22. n errors are automatically detected and reported A coding error is a discrepancy between the waveform codes 150 a E435 T60401F date 07 25 97 experimenter WAvG experiment PROBING file Total Number Total Probing Time Average Duration Percent of Time species Ap treatment Vf insect no location morph custom 2 custom 3 remarks Dur Avg Dur sec Waveform sec 44 8 26 9 257 2 31 1 982 4 baseline 268 5 1 161 5 1286 1 62 2 1964 7 COON Oa h OND Oo 100 0 3743 0 178 2 ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA 6 probes 3474 5 sec 579 1 sec 92 8 Second generation after start of clonal colony Vol 91 no 1 b VI Statistics duration lt T60401F T60401F T60401F T60401F T60401F T60401F T60401F T60401F T60401F 760401F T60401F T60401F 160401F T60401F T60401F T60401F T60401F T60401F T60401F T60401F T60401F 34 6 65 1 289 5 309 6 947 4 1019 9 1041 7 1178 5 1226 1 1255 1 1465 3 1484 6 1523 4 2045 2 2087 3 3414 2 3480 0 3499 3 3528 0 3550 1 3743 0 al vir al vir al vir al vir al vir al vir al vir alvir al vir al vir al vir al vir al vir al vir al vir al vir al vir al vir al vir al vir al vir s o E 2E Vv D4 D4 D4 D4 D4 D4 D4 D 4 D4 D4 D4 D4 D4 D4 D4 D4 D4 D4 D4 D4 D4 DADA RE RWW WN NY o lt probe no HON STN kK OTN OTN ww ON x oa gt NN I
23. ng behaviors McLean and Kinsey 1967 1968 Tjallingii 1978 1985 Postacquisition analysis of EMIF data formerly was rather labor intensive because waveform dura tions had to be measured manually from paper print outs A number of researchers have developed soft 146 ware to record and analyze EMIF data by computer Noyes et al 1990 wrote data acquisition and anal ysis software in BASIC C and 8086 assembly lan guage under MS DOS for use with AC EMIF sys tems For use with DC EMIF systems Stylet versions 1 0 2 0 and 2 3 Tjallingii and Mayoral 1992 and MacStylet Febvay et al 1996 were written in ASYST Scientific toolbox software Keithley Met rabyte Taunton MA and QuickBasic respec tively An upgraded version of MacStylet written in C and adapted for DOS and PowerPC systems is under development Febvay et al 1996 Finally Gruenhagen and Backus 1996 designed EMIF data acquisition and analysis software for AC EMIF sys tems using Streamer Keithley Metrabyte Reflex and SuperKey Borland Scotts Valley CA Researchers so far were forced to write their own data acquisition and data analysis software because professional data acquisition software was too ex pensive and typically lacked the analysis functions required for this particular application To over come this obstacle we developed an EMIF analysis method using commercially available professional data acquisition software in conjunction wit
24. ological Society of America Lanham MD McLean D L and M G Kinsey 1964 A technique for electronically recording aphid feeding and salivation Nature Lond 203 1358 1359 1965 Identification of electrically recorded curve pat terns associated with aphid salivation and ingestion Nature Lond 205 1130 1131 1967 Probing behaviour of the pea aphid Acyrthosiphon pisum I Definitive correlation of electronically re 154 corded waveforms with aphid probing Ann Entomol Soc Am 60 400 406 1968 Probing behaviour of the pea aphid Acyrthosiphon pisum II Comparison of salivation and ingestion in host and non host plant leaves Ann Entomol Soc Am 61 730 739 Microsoft 1993 Microsoft Excel version 5 0 user s man ual Microsoft Redmond WA 1995 Getting results with Microsoft Office for Windows 95 version 7 0 Microsoft Redmond WA 1995 1996 Getting results with Microsoft Office 97 Mi crosoft Redmond WA Montllor C B and W F Tjallingii 1989 Stylet pene tration by two aphid species on susceptible and resis tance lettuce Entomol Exp Appl 52 103 111 Noyes S D C Margolies J C Reese A Dixon and P J Bramel Cox 1990 Computer acquisition and analysis of insect feeding monitor data Ann Entomol Soc Am 83 1224 1231 Powell G R Harrington and N J Spiller 1992 Stylet activities and potato virus Y vector efficiencies by the aphids Brachycaudus helichrysi and Drepano
25. ons in Entomology Proceedings Entomological Society of America Lanham MD VAN GIESSEN AND JACKSON ELECTRONICALLY MONITORED FEEDING BEHAVIOR 153 Backus E A and W H Bennett 1992 A new AC elec tronic insect feeding monitor for fine structure analysis of waveforms Ann Entomol Soc Am 85 437 444 Blua M J and T M Perring 1992 Effects of zucchini yellow mosaic virus on colonization and feeding be havior of Aphis gossypii Homoptera Aphididae ala tae Environ Entomol 21 578 585 Caillaud C M J S Pierre B Chaubet and J P Di Pietro 1995 Analysis of wheat resistance to the cereal aphid Sitobion avenae using electrical penetration graphs and flow charts combined with correspondence analysis Entomol Exp Appl 75 9 18 Calatayud P A Y Rahb W F Tjallingii M Tertuliano and B Le Ru 1994 Electronically recorded feeding behaviour of cassava mealybug on host and non host plants Entomol Exp Appl 72 219 232 Calderon J D and E A Backus 1992 Comparison of the probing behaviors of Empoasca fabae and E krae meri Homoptera Cicadellidae on resistant and sus ceptible cultivars of common beans J Econ Entomol 85 88 99 Cole R A 1994 Locating a resistance mechanism to the cabbage aphid in two wild Brassicas Entomol Exp Appl 71 23 31 Dataq Instruments 1990 Advanced CODAS user s man ual software release level 3 manual revision C Dataq Instruments Akron OH
26. re the default setting for calculation is manual This means that after all appropriate changes have been made to the workbook the user should press function key F9 to ANNALS OF THE ENTOMOLOGICAL SOCIETY OF AMERICA Vol 91 no 1 View after pressing Show Errors button Fig 3 Example of worksheet III of the Excel work book for the analysis of electronically monitored insect feeding behavior By pressing 1 of 2 buttons A the user can import data files with various formats see Fig 1 C F and G In addition data can be typed directly into this worksheet or data can be cut and pasted from other Excel files or other applications After importing a data file waveforms that are not specified in worksheet II are marked with an x coding error and waveform codes that occur twice in a row are indicated by a dupli cation error B Pressing the button labeled Show Er rors C will reduce the visible part of the data file to just the detected coding and duplication errors D to facili tate correction of long data files Pressing the button la beled Show Al will again show the complete data file instruct Excel to recalculate all formulas On faster computers the user can change calculation to au tomatic by pressing a button on worksheet II Fig 2E A custom EMIF workbook can be created e g tailored to a particular insect species or experiment by saving a version of the workbook
27. s the user can select from 5 different zoom factors on each worksheet Fig 2G Each worksheet also has a gt gt and a lt lt button Fig 2H to page forward and backward through the workbook Even though these buttons are present on all 7 worksheets they are shown only for worksheet II in Fig 2 Each worksheet is de scribed in detail below January 1998 VAN GIESSEN AND JACKSON ELECTRONICALLY MONITORED FEEDING BEHAVIOR 147 WinDaq EX Microsoft Excel EMIF Workbook Instructions Il Specifics see Fig 2 Windaq Notepad File WNF oe 0 00E 00 o 6 00 00 008 00 0 00 00 o 00e 00 3 466 0 0 0 006 00 0 00e 00 0 ode 00 0 0000 BSED 0 a40e o1 7 ERTED ES ce ID 2a0e 02 o J 6 10E1 4 60E Ory1 76E 01 2006 51 3 i0E 02 0 107E 00 4 Yee o2 2 20E2 zee OH e a7e 02 0 8206 01 6506 01 7226 01 2 08 01 1026 03 o 1 87E 00 4 306 02 7 616 02 2 05E D7 KEDERE EEA memo o roon 290e 01 1866 07 3086 07 Or CE Eac EE ET BaL h 1019 9 s 1017 z 1178 5 Fig 1 Diagram of a new method to analyze recordings of electronically monitored insect feeding behavior EMIF Waveforms are identified by the user using WinDaq EX postacquisition software A The data cursor B is moved across the signal to the transition between 2 successive waveform occurrences and the Notepad File utility is used to save a code that specifies the type of waveform follow
28. siphum platanoidis Entomol Exp Appl 62 293 300 Powell G J Hardie andJ A Pickett 1993 Effects of the antifeedant polygodial on plant penetration by aphids assessed by video and electrical monitoring Entomol Exp Appl 68 193 200 Rahb Y G Febvay B Delobel and G Bonnot 1998 Amino acids and proteins as cues in aphid plant inter actions In G P Walker and E A Backus eds Prin ciples and applications of electronic monitoring and other techniques in the study of homopteran feeding behavior Thomas Say Publications in Entomology Proceedings Entomological Society of America Lan ham MD in press Reese J C D C Margolies E A Backus S Noyes P J Bramel Cox and A G O Dixon 1994 Characteriza tion of aphid host plant resistance and feeding behavior through use of a computerized insect feeding monitor pp 52 72 In M M Ellsbury E A Backus and D L Ullman eds History development and application of AC electronic insect feeding monitors Thomas Say Publications in Entomology Proceedings Entomolog ical Society of America Lanham MD SAS Institute 1989 SAS STAT user s guide version 6 4th ed SAS Institute Cary NC Saxena R C and Z R Khan 1985 Electronically re corded disturbances in feeding behavior of Nephotettix virescens Homoptera Cicadellidae on neem oil treated rice plants J Econ Entomol 78 222 226 Shuckle R H and C Quiroz 1994 Use of the ele
29. sly the time needed to recognize and mark various waveforms will depend on the quality of the recording wave form variability the number of waveforms to iden January 1998 tify the level of experience of the user and com puter speed The time spent analyzing recording A using the EMIF workbook can be subdivided as follows 1 60 s to fill out the specific experimental variables in worksheet II this will take less time when a series of similar files are being analyzed 2 between 20s using a 200 MHz Pentium Pro PC and 45 s using a 66 MHz Pentium PC to import the data file into the workbook 3 10 20 s to rename and save the workbook to floppy or hard disk depending on the computer used This means that the complete anal ysis of recording A took between 1 and 2 min The analysis of recording B 351 waveforms took only slightly longer Reliable pattern recognition software to charac terize waveforms automatically has yet to be devel oped Tjallingii 1988 and Rahb et al 1998 in vestigated whether signal analysis through Fourier processing could lead to automated recognition but they were not successful The major reason for this lack of success is the enormous variability inherent to EMIF data Tjallingii 1988 No doubt advances in pattern recognition will facilitate this in the fu ture we are exploring new approaches However until reliable pattern recognition software is devel oped the human eye is super
30. workbook can be tailored to the different sets of waveforms generated by various homopteran spe cies e g aphids planthoppers leafhoppers white flies psyllids In addition the user can use codes of up to 5 characters long Data File Compatibility The EMIF workbook can be used in conjunction with all data acquisition software capable of exporting waveform codes and cumulative time and all data files with a 2 column format column 1 waveform code column 2 cu mulative time see Fig 1 The same format can be achieved if waveform codes and transition times are compiled by hand EMIF System Compatibility Originally the EMIF workbook was developed for use with WinDaq and the Missouri AC insect feeding monitor Backus and Bennett 1992 However the workbook can be used for the analysis of EMIF recordings made with any other data acquisition software with signals from either AC or DC systems Speed of Analysis We timed the steps involved in the analysis of recordings A A gossypii on C au rantifolii duration 4 h and B E kraemeri on P vulgaris duration 2 8 h Fig 7 Using WinDaq s Notepad Utility the identification and separation of the 73 waveform occurrences contained in record ing A took 5 min and 40 s whereas it took 30 min to identify and separate the 351 waveforms occur rences in recording B This means that it took ap proximately 5 s to identify a single waveform oc currence and save it to file Obviou
Download Pdf Manuals
Related Search