Selection of rice accessions
In September 2014 and 2015, we collected plants just before harvest in two villages in YYT i.e., Jingkou and Xiaoshuijing (Supplementary Information SI1 Table SI1.1) from the six most popular rice landraces in the area (20 diseased plants + 10 healthy plants/ field, two fields / variety in each village). Panicles with mature seeds were kept for all plants so that each plant accessions could be selfed and grown again for DNA extraction and/or for multiplication. From this sampling, a set of 92 rice accessions representing indiginous landraces (Table SI1.1), was genotyped using Genotyping-by-Sequencing (Arbealz et al., 2015). DNA for these 92 rice accessions was extracted from individual plants. Half a leaf of each plant was ground into powder in liquid nitrogen. A volume of 750 ul of pre-warmed extraction buffer (CTAB 2% w/v, Tris-HCl 200 mM pH 8.0, EDTA 20 mM pH 8.0, NaCl 1.4 M, Polyvinylpyrrolidone (K30) 1% w/v, β-mercaptoethanol 1% v/v) was added to the powder and incubated at 65°C for 45 min. After centrifugation 15 min at 13000 rpm, the supernatant was recovered and extracted with the same volume of dichloromethane: isoamyl-alcohol (24:1). After centrifugation 15 min at 13000 rpm, the resulting supernatant was treated with RNAse A (0,1 ug/ml final) for 20 min at 37°C. The remaining nucleic acids were precipitated with cold isopropanol for 20 min at -20°C and centrifuged 10 min at 15000 rpm. The DNA pellet was washed once with ethanol 76%:sodium acetate 200 mM, once with ethanol 76%: sodium acetate 10 mM and finally resuspended in TE (Tris-HCl 10mM, EDTA 1mM pH8) buffer. DNA quality and quantity were checked using Nano-drop, Qubit® dsDNA BR Assay Kits and on agarose gel. Library preparation and sequencing were performed at UMR AGAP (Montpellier, France) following the description in (Elshire et al., 2011). DNA was digested with ApeKI for library preparation and subsequent sequencing with an Illumina Genome Analyzer II (San Diego, California, USA).
To evaluate the genetic distance between YYT landraces and worldwide representatives of various rice sub-species, we selected 113 accessions of the worldwide rice sequencing data studied in Huang et al. (2012) and 103 accessions from the study of Wang et al. (2017) from the European Nucleotide Archive (ENA) database, chosen to maximize the geographic origins, sub-species representatives and genetic diversity (Supplementary information SI1).