AraPheno: a public database for Arabidopsis thaliana phenotypes

Abstract

Natural genetic variation makes it possible to discover evolutionary changes that have been maintained in a population because they are advantageous. To understand genotype-phenotype relationships and to investigate trait architecture, the existence of both high-resolution genotypic and phenotypic data is necessary. Arabidopsis thaliana is a prime model for these purposes. This herb naturally occurs across much of the Eurasian continent and North America. Thus, it is exposed to a wide range of environmental factors and has been subject to natural selection under distinct conditions. Full genome sequencing data for more than 1000 different natural inbred lines are available, and this has encouraged the distributed generation of many types of phenotypic data. To leverage these data for meta analyses, AraPheno (https://arapheno.1001genomes.org) provide a central repository of population-scale phenotypes for A. thaliana inbred lines. AraPheno includes various features to easily access, download and visualize the phenotypic data. This will facilitate a comparative analysis of the many different types of phenotypic data, which is the base to further enhance our understanding of the genotype-phenotype map.

Figure 1.

Screenshot of the detailed view for a phenotype of interest (https://arapheno.1001genomes.org/phenotype/43/). (A) General information such as ‘Scoring’ or various ontology terms are displayed in text form. (B) The geographic distribution of the samples that were scored are displayed as a GeoChart. (C) A powerful Explorer widget relates the phenotype value of each sample to its geographic location, thus bringing out potential geographic patterns.

Figure 2.

Screenshot of the detailed view for a specific accession (https://arapheno.1001genomes.org/accession/6909/). (A) General information such as ‘Country’ or ‘Collector’ are displayed in text form. (B) A map shows the geographic origin of the accession. (C) Various aggregated statistics about the ontologies for the (D) list of phenotypes that the accession was scored in.

Figure 3.

Screenshot of the Phenotype-Correlation results (https://arapheno.1001genomes.org/correlation/6,29,30,31,49,102,99,53,86,39/). (A) User can specify the correlation method (Pearson and Spearman are supported). (B) The Phenotype–Phenotype Correlation Plot displays pairwise correlation values for the selected phenotypes. When the user moves the mouse over a cell the (C) The Phenotype–Phenotype Scatter Plot plots the corresponding phenotypic values against each other and the (D) The Phenotype Sample Overlap Diagram shows the overlap between the two selected phenotypes.