Phytoplasmas are a group of cell wall-less bacteria with small genome sizes (530-2220 kb) belonging to the class Mollicutes. They are obligate parasites of plant phloem tissue and are transmitted from plant to plant by sap-feeding insect vectors of the order Hemiptera (Weintraub and Beanland 2006). They are responsible for hundreds of plant diseases worldwide, causing serious damage to a variety of cultivated and wild plants (Bertaccini et al. 2014). Phytoplasmas were observed for the first time in 1967 (Doi et al. 1967) and only recently classified based on phylogenetic analysis of 16S rRNA and ribosomal protein (rp) gene sequences, revealing their diversity and phylogenetic position (Lee et al. 2000; IRPCM, Phytoplasma/Spiroplasma Working Team - Phytoplasma Taxonomy Group 2004). To date, 33 groups and more than 120 subgroups have been recognised based on 16Sr genetic data and 44 taxa have been given the provisional taxonomic status of ‘Candidatus’ (Harrison et al. 2018), denoting not-yet cultured bacteria. As phytoplasmas do not satisfy the four Koch postulates (i.e. four criteria used to determine if an organism is the causal agent of a disease), indirect biological evidence, such as electron microscopy observation, phytoplasma and symptom elimination after tetracycline treatments (Ishiie et al. 1967), insect and dodder transmission, has confirmed association of phytoplasmas with numerous plant diseases worldwide (Bertaccini 2007; Hogenhout et al. 2008).

The only known phytoplasma vectors are hemipteran insects belonging to the suborder Auchenorrhyncha (Fulgoromorpha and Cicadomorpha) and Sternorrhyncha (family Psyllidae). An insect acquires the phytoplasma from an infected plant, then the phytoplasma must overcome two major barriers (insect gut and salivary glands cells) in order to reach the saliva for subsequent introduction into another plants' phloem (Ammar and Hogenhout 2006). An insect species is qualified to be a vector if it is competent to transmit the pathogens (vector competence) and this condition is related to the possibility of the phytoplasma to overcome the above-mentioned barriers. The screening of hemipteran species to confirm their vector competence is usually performed using traditional laboratory methods that are extremely time-consuming and expensive and, in general, have only been performed in cases where the pathogen causes severe economic losses to agriculturally important plants. As a consequence, basic knowledge of insect vector ecology and epidemiology of the plant diseases caused by phytoplasmas remains highly incomplete and vectors of many phytoplasmas still have not been identified. In order to integrate and organise data available from disparate sources, improve knowledge of global patterns of Hemiptera, phytoplasma and plant relationships and facilitate research, an open source database platform was created. This paper presents a complete searchable online database of Hemiptera-Phytoplasma-Plant (HPP) biological interactions worldwide. The database has large potential impacts on a variety of research disciplines ranging from systematics to invasive species ecology and vector biology. It will facilitate synthesis of all available and relevant data on Hemiptera-Phytoplasma-Plant associations and testing of ecological and evolutionary hypotheses.

Individual records in the raw-HPP database consist of specific pairwise interactions detected in specific countries with their literature citations. In the database, the type of relationship with phytoplasma was defined by a positive detection in an insect body or plant tissues with a consequent assignment to the phytoplasma group or subgroup (e.g. 16SrI or 16SrI A) or to a not yet characterised phytoplasma. Investigated relationships that resulted in negative detection of a phytoplasma are reported as “not detected”. The record was included if the positive or negative detection was proven using microscopic and serology-based techniques, PCR targeting the 16S–23S rRNA genes, observed or virtual RFLP fingerprinting or DNA sequencing methods (Smart et al. 1996; Namba 2011).

The 3I-HPP web interface summarises the history of the investigation for each insect and plant species. Therefore, the web interface includes 5 different categories for each taxon level, 3 for insects and 2 for plants. The category assigned to each taxon may change quickly as new investigations are carried out and new data are added to the database.

For each insect species (taxon page), one or more of the three possible categories were assigned, according to the available evidence about the capability to transmit the phytoplasma, as follows:

1. Vectorial Competence. A hemipteran species is considered a COMPETENT VECTOR for a certain phytoplasma if the capability of the latter to overcome the barriers of gut and salivary glands of the insect has been proven using classical acquisition/inoculation experiments in the laboratory or inoculation trials with caged infected specimens collected from the field. Moreover, the successful transmission of the phytoplasma to the plant has been detected on tested insects and plants using standard molecular methods, in addition to detection of clear symptoms due to phytoplasma presence in the plant. A less strict evaluation may be applied for papers published before the PCR molecular era (before 1980s), especially if the same insect species were later demonstrated to be competent vectors in the same region.
2. Positive Detection in insect. A hemipteran is considered a POTENTIAL VECTOR if the phytoplasma has been detected in its body using standard molecular methods. Status as a potential vector does not imply the ability to transmit the phytoplasma from plant to plant. Such insects may represent dead-end hosts for the phytoplasma if it is later shown that they are not competent as vectors.
3. Negative Detection in insect. A hemipteran is considered a POTENTIAL NON-VECTOR if it tested negative for the presence of phytoplasma. The more times specimens of a species have tested negative, the more likely the species is a non-vector.For each plant species, two categories were defined according to the available evidence on detection of phytoplasma in the plant phloem, as follows:
4. Positive Detection in plant. A plant is considered INFECTED BY phytoplasma if the pathogen has been successfully detected in its tissues using standard molecular methods.
5. Negative Detection in plant. A plant is considered a POTENTIAL UNSUITABLE HOST if it tested negative for the presence of phytoplasma. The more times samples of a species have tested negative, the more likely that species is an unsuitable host for the phytoplasma.

I would like to thank Christopher H. Dietrich and Dmitry A. Dmitriev for their valuable support and constant feedback on the dataset and data paper. I also thank Richard Flood for his kind help and technical support. This study was funded by U.S. National Science Foundation grant (DEB – 16-39601).