Database on apple fruit pests of the EU to support pest risk assessments

Abstract

In December 2013, EFSA received a first mandate from the European Commission DG SANTE to gather information on the pests of apple fruit (Malus domestica ) in the EU territory (M‐2014‐0016). To satisfy the mandate, EFSA developed an overall approach to systematically collect information on EU apple pests and to organise it within a bespoke database with support from the Austrian Agency for Health and Food Safety. Test data were collected for 12 apple pests (6 insects and 6 pathogens). Based on the experience gathered, the initial database structure was adapted and refined by EFSA , permitting more efficient data gathering. In September 2017, as a follow‐up to the original mandate, EFSA was requested by the European Commission DG SANTE to test the suitability of the revised database in supporting risk assessors in third countries to carry out pest risk assessments of apple fruit as a commodity (M‐2017‐0203). As a first step, the data set on the 12 pests was migrated into the newly revised database structure. This was then converted into the MicroStrategy platform to provide a user‐friendly interface for data search and visualisation. At the same time, a new data entry tool using the systematic literature review software DistillerSR was created to enhance data extraction for future data collections. The interactive data reports were shared with the Canadian Food Inspection Agency (CFIA ) and the Ministry for Primary Industries (MPI ) in New Zealand for their testing and feedback as potential future users of the database. The overall feedback collected from CFIA and MPI confirms that the EU database on apple fruit pests could become an important tool to provide third countries the necessary technical and biological information for their pest risk assessments. Addressing feedback from CFIA and MPI has further improved the database structure and metadata. The database of apple pests can be included in the EFSA Scientific Data Warehouse and extended to provide a comprehensive list of pests and host plants.

1 Introduction

This report presents the structure and metadata of an updated concise database on pests of the apple fruit developed by the European Food Safety Authority following two mandates from the European Commission Directorate‐General for Health and Food Safety (EC DG SANTE).

This work is to provide a better framework facilitating and enhancing third countries’ risk assessments of pests of apple fruit in the EU which also provides for both better plant health protection and improved trade opportunities with third countries.

1.1 Background and Terms of Reference as provided by the requestor

1.1.1 Background

In December 2013, EFSA received a first mandate from the European Commission DG SANTE to gather information on the pests of apple fruit (Malus domestica ) in the EU territory (M‐2014‐0016). To satisfy the mandate, EFSA developed an overall approach to systematically collect information on EU apple pests and to organize it within a bespoke database with support from the Austrian Agency for Health and Food Safety, including: 1) the establishment of a comprehensive database structure; 2) extensive literature searches first to compile the list of pests of the apple fruit in the EU, and then to collect relevant data and populate the database. The EU database on apple fruit pests is modular and has a hierarchical organization, which makes it possible to adapt it to different data collection requests, choosing the level of detail that is required for the specific purpose.

An extensive search of literature published from 1950 to 2014 generated a list of 277 apple pests from around the world, 235 of them are present in the EU territory; 66 of them found to have a potential impact on apple fruit. The pests not considered to be relevant were not included, either because they had no direct effect on the apple fruit or they were considered as worldwide present, or they were not reported to be present in the EU territory in the time period 1950–2014, or for several of these reasons. For 12 of these pests11 Aphis pomi , Apple scar skin viroid, Archips rosanus , Cydia lobarzewskii , Eccopsia effractella , Elsinoe pyri , Hoplocampa testudinea , Monilinia fructigena , Neofabraea alba , Neofabraea perennans , Phytophthora syringae , Rhynchites aequatus .
(selected to cover a broad variety of pest types/groups), scientific and technical/grey literature was reviewed, and the data extracted were inserted in the initial pilot database version. Based on the experience gathered with these first 12 apple pests, the initial database structure was further adapted and refined by EFSA, permitting a more efficient data gathering. The pilot database on the 12 selected EU apple pests and the newly revised database structure have been shared and discussed by EFSA with European Commission DG SANTE services.

1.1.2 Terms of Reference

In September 2017, as a follow‐up to mandate M‐2014‐0016, EFSA was requested by EC DG SANTE to test the suitability of the revised database to support risk assessors to carry out pest risk assessments (M‐2017‐0203). This test should be done together with cooperating risk assessment organisations in Third countries (Canadian Food Inspection Agency (CFIA) and the Ministry for Primary Industries (MPI) in New Zealand, following previous interactions on this project) using the dataset collected for 12 pests in the frames of the first mandate. As the feedback from the cooperating Third countries is expected on the pilot 12‐pests dataset already collected, EFSA was requested, prior to this new exercise, to migrate the 12‐pests dataset into the newly revised database structure.

As a result of the feedback to be provided from the cooperating Third countries on the revised 12‐EU apple pests’ database, EFSA was requested, when applicable, to further improve the structure and metadata of the database and make then the revised 12‐EU apple pests’ database available to Commission and Members States.

After the successful completion of this validation phase, EFSA can be further requested with a follow up mandate by Commission to collect data and information on the remaining potentially relevant pests in the apple fruit present in the EU, prioritising the pests of relevance for EU trade with cooperating countries such as Canada.

2 Data and methodologies

2.1 Data management

The analysis of the outcome of the first mandate revealed that the process of data collection and management can be enhanced by a few procedural and technical developments explained in the sections below.

2.1.1 Development of a data entry tool in DistillerSR

During the first mandate, AGES experienced difficulties in dealing with an Excel‐based tool for the data extraction process, both due to the complex structure of the first data model, and due to the lack of co‐editing functionality.

In order to improve the process of data entry, the systematic review software DistillerSR22 https://www.evidencepartners.com/
was tested as an interface for literature review and data extraction. Successful tests using DistillerSR forms provided proof of concept, and review questions were prepared in line with the data model so that they best serve future data extraction (Appendix A).

2.1.2 Revised data model

The database on EU apple fruit pests was re‐designed to create a standardised data model for storing information related to the occurrence of pests on host plants (any pest on a chosen host, or a chosen pest an any host), and the impacts as a result of the occurrence, as observed with or without the application of phytosanitary and/or pest control measures.

2.1.2.1 Description of data model

The data model is connected to nearly 30 standard terminology catalogues that enable data collection without the use of free text which is a prerequisite of a queriable database. Although, according to the first mandate, the database was supposed to serve data collection on apple fruit pests, EFSA established the data model in such a way that it can facilitate future data collections for any plant pest or plant commodity. Data collection for a new pest or commodity may necessitate the update of some standard terminology catalogues.

The conceptual data model (Figure 1) does not specify single data elements. Data have been provided within a de‐normalised data table and for this reason all the entities can be considered tied to the Observation , and its Primary Key (recordId ). The Measures information section allows the reporting of multiple values (more measures can be applied at the same time) and the efficacy is generally reported as referred to all the measures applied. In the logical diagram, this is captured by the Measures entity which is logically linked to the Symptoms entity to make it evident that a measure cannot be reported if the impact is not specified.

The implemented data collection on the 12 test pests from the first mandate did not enforce the conceptual data model requirements (as presented on Figure 1). This is due to the fact that data were initially collected using a different test data model and some of the links required by the revised data model were not included in the initial test data model. All future data collections will be implemented following the new conceptual data model.

Figure 1

Conceptual data model of the EFSA database on apple fruit pests

The revised data model could serve the storage at EFSA of all data related to pests on host plants, including all relevant information that can be extracted from the article/publication and that provide the context for the observed occurrence, such as:

• pest species, life stage, form, development of the pest (pest identification );
• host species, part of the plant, cultivation, storage (host identification );
• related vectors (vector identification );
• geographical information (pest occurrence );
• symptoms and impacts (impact information );
• efficacy of applied phytosanitary and/or control measures (measures information ).

The data model (Table 1) has different sections dedicated to information on the pest, host, vector (if applicable), occurrence, impacts and measures. Within the sections, each line represents a single data element, for which data can be collected.

Table 1. Structure of the data model divided into sections and the related data elements within each section

Extracted data	Description
General information	In this section, general information about the information source is reported
recordId	Unique Id of the record
refId	Id of the citation as exported by EndNote/Distiller (all the information related to the publication can be retrieved and dynamically added to the data model)
informationId	Unique code identifying the different extraction records within the same Publication (Reference ID)
Pest identification	In this section, biological information on the pest is reported
pestSectionId	Unique Id of the section containing info about the pest
pestCode	EPPO code of the pest
pestLifestage	Life stage of the pest
stagesNumWithinLifestage	Total number of larval or nymphal stages
pathogenForm	Form of the pathogen
Overwintering	This flag can be referred either to the pestLifestage or to the pathogenForm
pestRegulatoryStatus	Regulatory status of the pest in Commission Implementing Regulation 2019/2072
pestListingEPPO	EPPO listing of the pest
pestHostDimension	Host range
pestOtherHosts	Hosts other than the commodity ‐ The field allows multiple values
pestDevelopmentAttribute	Key attributes of pest development
pestDevelopmentValueUnit	Unit to express the chosen attribute of development
pestDevelopmentValue	Value of chosen attribute
Host identification	In this section, information on the host is reported
hostSectionId	Unique Id of the section containing info about the host
hostCode	EPPO code of the host
hostPart	Part of the hostplant (where the pest was found and/or which shows symptoms)
hostProduction	Method of cultivation of the host
hostStorage	Method of storage of the host
Vector identification	In this section, information on the vector if applicable is reported
vectorSectionId	Unique Id of the section containing info about the vector
vectorCode	EPPO code of the vector
vectorLifestage	Life stage of the vector of the pest
vectorForm	Form of the vector
vectorDevelopmentAttribute	Key attributes of vector development
vectorDevelopmentValueUnit	Unit to express the chosen attribute of development
vectorDevelopmentValue	Value of chosen attribute
vectorTransmissionType	Mode of transmission of the pest by the vector
Pest occurrence	In this section, date and geographic information are reported related to the occurrence, as well as data on abundance when available
occurrenceSectionId	Unique Id of the section containing info about the occurrence
Country	Country of occurrence
Region	Region of occurrence (NUTS 2016 classification)
regionTxt	Region of occurrence name
Longitude	Longitude
Latitude	Latitude
Altitude	Altitude
Landcover	Land cover information of the place of occurrence
Landuse	Land use information of the place of occurrence
occurrenceYear	Year of occurrence
occurrenceMonth	Month of occurrence
occurrenceDay	Day of occurrence
pestPreferenceForHost	Host preference of the pest
qualAbundance	Abundance of the pest expressed in qualitative terms
quantAbundanceUnit	Unit to express the abundance of the pest in quantitative terms
quantAbundanceValue	Value of abundance
placeOcc	It can be used to report the place of occurrence of the pest
Impact information	In this section, symptoms and impacts on the host are reported related to the pest occurrence
symptomSectionId	Unique Id of the section containing info about symptoms and impacts
Symptom	Symptoms of pest damage ‐ The field allows multiple values
impactType	Type of loss
evaluationType	Type of evaluation of loss (quality / quantity / yield)
qualEvaluation	Evaluation of loss in qualitative terms
quantEvaluationUnit	Unit to express loss in quantitative terms
quantEvaluationValue	Value of loss
weatherContextCondition	Weather anomaly (reported at the time of occurrence) ‐ The field allows multiple values divided by ‘$’
Measures information	In thus section, information (qualitative and/or quantitative) on phytosanitary and control measures is reported
measureSectionId	Unique Id of the section containing info about the measures
phytosanitaryMeasure	Phytosanitary measures applied ‐ The field allows multiple values
pmQuantEfficacyUnit	Unit for the quantitative efficacy of the applied phytosanitary measures
pmQuantEfficacyValue	Quantitative efficacy of the applied phytosanitary measures
pmQualEfficacy	Qualitative efficacy of the applied phytosanitary measures as described in the source
pmQualEfficacyScale	Scale for the evaluation of the qualitative efficacy of the applied phytosanitary measures as described in the source
controlMeasure	Control measures applied ‐ The field allows multiple values.
contMeasParamCode	In case of chemical treatment, the field allows to specify which are the substances used for the treatment (multiple values can be provided)
cmQuantEfficacyUnit	Unit for the quantitative efficacy of the applied control measures
cmQuantEfficacyValue	Quantitative efficacy of the applied control measures
cmQualEfficacy	Qualitative efficacy of the applied control measures as described in the source
cmQualEfficacyScale	Scale for the evaluation of the qualitative efficacy of the applied control measures as described in the source
controlMeasuresContext	Context of control measure application
quantJointEfficacyUnit	Unit for the overall quantitative efficacy of the applied measures (both phyto and control)
quantJointEfficacyValue	Overall quantitative efficacy of the applied measures (both phyto and control)
quantJointEfficacy	Overall qualitative efficacy of the applied measures (both phyto and control) as described in the source
quantJointEfficacyScale	Scale for the evaluation of the qualitative efficacy of the applied measures (both phyto and control) as described in the source

2.1.3 Data migration

Data on the 12 pests collected by AGES during the first mandate was migrated into the new data model by means of supervised Extract Transform Load (ETL) procedures and imported again into EFSA's Data Collection Framework (DCF)33 The EFSA Data Collection Framework (DCF) is a platform through which data are submitted to EFSA by data providers.
to be validated against EFSA reference terminology (EFSA catalogues).

Due to time and resource constraints, an inclusion of data into the EFSA Scientific Data Warehouse (S‐DWH)44 The EFSA Scientific Data Warehouse (S‐DWH) allows the publication, analysis and distribution, in different formats and at different levels of granularity, of data collected by EFSA.
is planned at a later stage.

2.1.4 Creation of interactive reports

During the first mandate, data were collected in a Microsoft Excel template that made the data extraction and subsequent query rather difficult.

Following migration of the data on the 12 pests into the revised database, in order to make data extraction and query user‐friendly, harmonised data were included into MicroStrategy55 https://www.microstrategy.com/
data cubes. MicroStrategy is a business intelligence platform that enables the creation of interactive dashboards and data reports.

Interactive reports were created within the framework of an outsourced project enabling data querying and data extractions for the testing exercise, and eventually for the end‐users of the database.

2.2 Data testing with cooperating Third Countries

As requested in the second mandate, the two cooperating Third countries’ organisations (Canada CFIA and New Zealand MPI) were contacted separately to seek their comments and proposals for the revised data model and the data collected on the 12 pests.

A web meeting was organised with risk assessors of CFIA on 28 June 2018, in order to present them the MicroStrategy reports. Following the web‐meeting, access to the MicroStrategy reports was granted to CFIA and their comments were received on 27 September 2018 (see Appendix B, Table B.1).

Due to the large difference in time zone, it was not possible to organise a web meeting with the Ministry for Primary Industries (MPI) in New Zealand, but instead they received an email with detailed explanation on the MicroStrategy reports on 26 October 2018, and access rights to the reports were granted. Their comments were received on 26 November 2018 (see Appendix B, Table B.2).

3 Results

3.1 Literature review and data extraction

The DistillerSR tool has been found a good candidate for supporting literature review and data extraction activities for upcoming data collections. Following the selection of relevant papers in the EndNote66 https://endnote.com/
tool, the papers are imported into the DistillerSR platform for detailed review and data extraction.

3.2 Data migration

The data migration activity highlighted the partial lack of links between some of the connected variables and indirectly confirmed ex post the need for a revised data model.

The creation of a new harmonised data model for the pests in plants domain constituted an important achievement in terms of standardisation and harmonisation at an EFSA level. For instance, the data model has been used as a starting point to build the data model for the collection of data about the host plants of Xylella fastidiosa .77 https://efsa.onlinelibrary.wiley.com/doi/epdf/ https://doi.org/10.2903/j.efsa.2018.5408
Having demonstrated its use in organising information of a high‐profile pest, future inclusion of the apple fruit pest database in the EFSA S‐DWH could benefit from the ETL procedures already implemented for the Xylella database.

3.3 Data testing with cooperating Third Countries and consequent update of the database

The comments received from CFIA and MPI allowed EFSA to identify areas of enhancement for the data model and for the MicroStrategy reports. The comments are listed in Appendix B (Tables B.1 and B.2), with EFSA responses provided regarding what action has been taken or is planned or the reason why a suggestion cannot be satisfied.

Although containing data on only 12 pests, in general, the pilot database was appreciated by the commentators who perceived it as a useful tool for risk assessors upon its population with detailed data. The most valued features of the database were:

• The database provides access to the source behind the data, that is the user can see the metadata of the paper including the abstract,
• The sources involve not only scientific but also grey literature which is valuable particularly for third countries who may have poor or no access to EU grey literature, though some important data, e.g. on control measures, are often published in national databases and grey literature,
• The database may facilitate quick assessments on pest importance.

Considering the proposals for further improvement, some ideas have been implemented (see Appendix B) while others will be implemented when the database is migrated into EFSA's S‐DWH (e.g. complete taxonomic information, synonyms of pests) or upon receipt of a further mandate (e.g. further population of the database, preparation of a detailed user manual).

4 Conclusions

The overall feedback collected from CFIA and MPI confirms that the database on apple fruit pests occurring in the EU could become an important tool to provide third countries the necessary technical and biological information they require for pest risk assessments.

5 Recommendation

EFSA recommends the inclusion of the data collected on apple fruit pests of the EU into the S‐DWH and the extension of the pilot database to a more complete list of pests and host plants.

Abbreviations

CFIA

Canadian Food Inspection Agency

DCF

Data Collection Framework

ETL

Extract Transform Load

MPI

Ministry for Primary Industries

S‐DWH

Scientific Data Warehouse

Appendix A – List of review questions in DistillerSR

1. Which are the relevant information for this specific record?

1. Pest/Pest life stage/Pathogen form
2. Host/Host part
3. Vector/Vector life stage/Vector form
4. Geographical occurrence
5. Impact / Symptoms
6. Applied measures

If Q1 is A. “Pest / Pest life stage / Pathogen form”

2. Life stage of the pest

3. Form of the pathogen

4. Regulatory status of the pest in Commission Implementing Regulation 2019/2072

5. EPPO listing of the pest

6. Host range

7. Key attributes of pest development

8. Unit to express the chosen attribute of development

9. EPPO code of the pest

10. Value of chosen attribute

If question 1 is B Host / Host part

11. EPPO code of the host

12. Part of the host plant (where the pest was found and/or which shows symptoms)

13. Method of cultivation of the host

14. Method of storage of the host

If Q1 is C. “Vector / Vector life stage / Vector form

2. Life stage of the pest

3. Form of the pathogen

4. Regulatory status of the pest in Commission Implementing Regulation 2019/2072

5. EPPO listing of the pest

6. Host range

7. Key attributes of pest development

8. Unit to express the chosen attribute of development

9. EPPO code of the pest

10. Value of chosen attribute

15. EPPO code of the vector

16. Life stage of the vector of the pest

17. Form of the vector

18. Key attributes of vector development

19. Unit to express the chosen attribute of development

20. Value of chosen attribute

21. Mode of transmission of the pest by the vector

If Q1 is D. “Geographical occurrence”

22. Country of occurrence

23. Geographical occurrence data availability

28. Landcover

29. Land use

30. Year of occurrence

31. Month of occurrence

32. Day of occurrence

33. Host preference of the pest

34. Abundance of the pest expressed in qualitative terms

35. Unit to express the abundance of the pest in quantitative terms

36. Value of abundance

If Q1 is E. “ Impact / Symptoms”

2. Life stage of the pest

3. Form of the pathogen

4. Regulatory status of the pest in Commission Implementing Regulation 2019/2072

5. EPPO listing of the pest

6. Host range

7. Key attributes of pest development

8. Unit to express the chosen attribute of development

9. EPPO code of the pest

10. Value of chosen attribute

11. EPPO code of the host

12. Part of the hostplant (where the pest was found and/or which shows symptoms)

13. Method of cultivation of the host

14. Method of storage of the host

37. Symptoms of pest damage

38. Type of loss

39. Type of evaluation of loss

43. Weather anomaly

If Q1 is F. “Applied measures”

2. Life stage of the pest

3. Form of the pathogen

4. Regulatory status of the pest in Commission Implementing Regulation 2019/2072

5. EPPO listing of the pest

6. Host range

7. Key attributes of pest development

8. Unit to express the chosen attribute of development

9. EPPO code of the pest

10. Value of chosen attribute

11. EPPO code of the host

12. Part of the host plant (where the pest was found and/or which shows symptoms)

13. Method of cultivation of the host

14. Method of storage of the host

37. Symptoms of pest damage

38. Type of loss

39. Type of evaluation of loss

43. Weather anomaly

44. Phytosanitary measures applied

45. Type of evaluation of efficacy of the phytosanitary measures

50. Control measures applied

51. Type of evaluation of efficacy of the control measures

56. Context of control measure application

57. Type of evaluation of overall efficacy of the applied measures (both phytosanitary and control)

Appendix B – Comments received from cooperating Third Countries

Table B.1. Comments received from the Canadian Food Inspection Agency and actions taken as a result

	Comments received	Actions taken
General comments	It is a fantastic database and EFSA has done a great job developing it	–
	Very useful product	–
	It is very slow to load	The MicroStrategy environment platform has been migrated to a cloud environment. This activity was already planned for the entire Scientific‐Data Warehouse and it will hopefully ensure a more reliable system maintenance. Since the migration, the data loads much faster
	If EFSA is able to populate it, CFIA will certainly be able to use it in their risk assessments	Further population is pending receipt of a new mandate from the European Commission
	Listing the references for each report is a major advantage to this database versus other similar databases	–
Specific comments	The amount of information is excellent, and the inclusion of papers published in local workshop proceedings (for example for G. lobarzewskii ) is very useful	–
	There is a lot of overlap in the information provided in tabs 1 and 5. By interchanging column “Country” and “Pest” in tab 1, you get the same information then found in tab 5. It is unlikely that the format under tab 5 be useful for our assessment. From our perspective, this tab could probably be deleted with minimal impact	Tab 5 has been removed
	It would be nice to be able to make a selection for pests at the genus level that would include all accounts at the genus level (identified to species and not). For example, right now if you select Neofabraea sp., it only shows accounts of Neofabraea that have not been identified at the genus level (i.e. “Neofabraea sp.”). It would be ideal if there was a search option (maybe for Neofabraea spp. or just for Neofabraea ) that would show all Neofabraea identifications (i.e. “Neofabraea alba ”, “Neofabraea perennans ”, and “Neofabraea sp.”).	Complete information about genus, species, subspecies of a pest will be available once data will be migrated to the EFSA Scientific Data Warehouse. Information about the pest will be codified using the EPPO codes (EPPOPEST EFSA catalogue)
	In Tab 2 and Tab 3, sometimes just the variety is listed without listing the species (e.g. just “var. Fuji” is listed instead of “Malus domestica var. Fuji”. This will become very confusing when species other than apple are added	Complete information about genus, species, subspecies of a host will be available once data will be migrated to the EFSA Scientific Data Warehouse. Information about the host will be codified against the EPPO codes (EPPOHOST EFSA catalogue)
	It would be nice if in tab 3 (and others) a filter option could be added to the columns (e.g. filter by plant part and/or symptom). Without this feature, some of the information in the columns is not very accessible and can only be accessed by scrolling through the full length of the column and waiting for it to load – I suspect this may become burdensome when the database grows	Filters by plant part and symptom have been added in Tab 3. Since at the moment the database has a relatively low amount of data, and the four filters are hierarchically related (that is if you select one, it will automatically sub‐select the other three), it may happen that with certain selections there will be no data visible. By clicking on the available values (including “All”), there should always be data visible
	The option for exporting data is very useful (references and others)	–
	A reference for the control measures in tab 4 would be useful	The references have been added in Tab 4. Tab 4 has been split into Tab 4a and 4b. Tab 4a shows the world map, while in Tab 4b the references behind the data are visible. The two sub tabs are not interconnected for the time being

Table B.2. Comments received from the New Zealand Ministry for Primary Industries and actions taken as a result

	Comments received	Actions taken
General comments	Overall, the database is very useful and generally easy to operate, though quite slow in this version The different options for displaying data would be a real time‐saver for refining searches to target search results more effectively	The MicroStrategy environment platform has been migrated to a cloud environment. With this, the data loads much faster
	The database would certainly facilitate the process of conducting IRAs or PRAs on EU commodities or pests if it was extensively populated	Further population is pending receipt of a new mandate from the European Commission
Specific comments	As above, the categorisation of information by geographic region, host, plant part etc. is helpful. For Plant parts, there is a category “Part not specified” but there doesn't seem to be one for the geographic region e.g. where the paper is about sequencing for identification/ cladistics etc and where no particular geographic data is included	As specified on Figure 1, data inserted in the database is tied to an observation (or report) and thus must have geographic information. When geographic information is not available, data cannot be recorded
	As well as detailed risk analyses, the database could be useful for making quick calls on the relative importance of various pests/pathogens. Tab 6 in particular (Pests geographic distribution) provides a very quick and easily understood high level indication of relative global importance	–
	For regulators, the tab with “Control measures taken by countries” is a very useful feature. It is difficult to get this information all in one place. However, it was unclear whether this information is referenced. If not referenced, it would be of less value. Even grey literature references are useful (in fact probably more so than data in abstracting databases or obtainable via Google searches)	The references have been added in Tab 4. Tab 4 has been split into Tab 4a and 4b. Tab 4a shows the world map, while in Tab 4b the references behind the data are visible. The two sub tabs are not interconnected for the time being
	It was unclear how to save a report, if that is possible in this demo version	In each table it is possible to export data in xlsx, csv or pdf formats
	What facility is there for incorporating synonyms? For example, the preferred name of Neofabraea alba is Phlyctema vagabunda . Incidentally this species is present in NZ (which the database doesn't flag)	At the moment, the database does not incorporate synonyms. Information about synonyms will be made available in the EPPOPEST EFSA catalogue and then in the reports once data will be migrated to the Scientific Data Warehouse
	It wasn't immediately obvious that it is necessary to filter the reference column by pest, so it's initially confusing as to why the number of reported occurrences (left hand column) doesn't sometimes match with the number of literature references in the right hand, however the navigation tabs at the bottom are great	A detailed user manual will be made available at a later stage
	It would be good to have the ability to resize the RHS panel with the abstracts in so the whole abstract can be read without scrolling across	The size of the columns can be changed but not the size of the text. We did the setting to automatically fit the available space. Within this setting, it is still possible to make a specific column wider so that a larger amount of text could be seen on the same page. As for the size of the text, the only option is to use the zoom of the browser which influences the entire page
	Is it possible to get rid of the box that results from a search in “Control measures”? It seems to persist when you move to another tab or even another search, and to be necessary to reload the database to escape it	MicroStrategy allows to add filters by selecting a value in a single cell. By selecting the same cell, the tool removes the filter