PhenoPlasm: a database of disruption phenotypes for malaria parasite genes

Abstract

Two decades after the first Plasmodium transfection, attempts have been made to disrupt more than 3,151 genes in malaria parasites, across five Plasmodium species. While results from rodent malaria transfections have been curated and systematised, empowering large-scale analysis, phenotypic data from human malaria parasite transfections currently exists as individual reports scattered across a the literature. To facilitate systematic analysis of published experimental genetic data across Plasmodium species, we have built PhenoPlasm ( http://www.phenoplasm.org), a database of phenotypes generated by transfection experiments in all Plasmodium parasites. The site provides a simple interface linking citation-backed Plasmodium reverse-genetic phenotypes to gene IDs. The database has been populated with phenotypic data on 367 P. falciparum genes, curated from 176 individual publications, as well as existing data on rodent Plasmodium species from RMgmDB and PlasmoGEM. This is the first time that all available data on P. falciparum transfection experiments has been brought together in a single place. These data are presented using ortholog mapping to allow a researcher interested in a gene in one species to see results across other Plasmodium species. The collaborative nature of the database enables any researcher to add new phenotypes as they are discovered. As an example of database utility, we use the currently available datasets to identify RAP (RNA-binding domain abundant in Apicomplexa)-domain containing proteins as crucial to parasite survival.

Keywords: Malaria; Plasmodium; database; genetic; knock-out; phenotype.

Figure 1.. The results of a search for ‘kinase’ genes, showing phenotype data, both from P. falciparum experiments and those in rodent models across multiple lifecycle stages.

Green ticks indicate mutant viability, and circled green ticks indicate wild-type phenotype. Red crosses indicate failure to disrupt the gene, and red exclamation marks indicate a phenotype different from wildtype. The icons are either shown in full opacity (indicating they apply to the gene in the species queried) or semi-transparent (indicating they refer to orthologous genes in other species).

Figure 2.. The phenotype page for the P. vivax CRK4 gene.

Though no experimental data is available directly from P. vivax, published results are shown from P. berghei and P. falciparum, with references to the original datasets from which likely data in P. vivax could be inferred. This gene is essential and has therefore been refractory to all attempts to disrupt it by classical reverse genetics, but a conditional system has also been recently applied in P. falciparum, allowing a more detailed phenotype to be assigned to the gene from our taxonomy.

Figure 3.. The number of genes with phenotyping data available in PhenoPlasm for each Plasmodium species, and the source of these annotations.

Figure 4.. A timeline of PubMed dates associated with publications reporting knock-out phenotypes for P. falciparum genes, from the first gene disruption in 1997 to mid-2017.

The values shown are cumulative from all previous years. Around 25 genes per year have had disruption attempts reported since the year 2000. The spikes that occur in 2008 and 2011 largely represent two individual publications systematically knocking out exported genes ( Maier et al., 2008) and kinases ( Solyakov et al., 2011) respectively.