Drone phenotyping and machine learning enable discovery of loci regulating daily floral opening in lettuce

Abstract

Flower opening and closure are traits of reproductive importance in all angiosperms because they determine the success of self- and cross-pollination. The temporal nature of this phenotype rendered it a difficult target for genetic studies. Cultivated and wild lettuce, Lactuca spp., have composite inflorescences comprised of multiple florets that open only once. Different accessions were observed to flower at different times of day. An F6 recombinant inbred line population (RIL) had been derived from accessions of L. serriola x L. sativa that originated from different environments and differed markedly for daily floral opening time. This population was used to map the genetic determinants of this trait; the floral opening time of 236 RILs was scored over a seven-hour period using time-course image series obtained by drone-based remote phenotyping on two occasions, one week apart. Floral pixels were identified from the images using a support vector machine (SVM) machine learning algorithm with an accuracy above 99%. A Bayesian inference method was developed to extract the peak floral opening time for individual genotypes from the time-stamped image data. Two independent QTLs, qDFO2.1 (Daily Floral Opening 2.1) and qDFO8.1, were discovered. Together, they explained more than 30% of the phenotypic variation in floral opening time. Candidate genes with non-synonymous polymorphisms in coding sequences were identified within the QTLs. This study demonstrates the power of combining remote imaging, machine learning, Bayesian statistics, and genome-wide marker data for studying the genetics of recalcitrant phenotypes such as floral opening time.One sentence summaryMachine learning and Bayesian analyses of drone-mediated remote phenotyping data revealed two genetic loci regulating differential daily flowering time in lettuce (Lactuca spp.).

AFLAP: Assembly-Free Linkage Analysis Pipeline usingk-mers from whole genome sequencing data

“…AFLAP generated nine linkage groups for each parent colinear with the 2.4 Gb genome assembly ( Figure 4) (13). A nine-linkage group, 1,711 cM nine-linkage group compound map (Additional File 3: Figure S1) was concordant with a 1,883 cM genetic map previously obtained via a conventional read alignment and variant calling workflow (12). Therefore, AFLAP can efficiently generate accurate genotype tables for linkage analysis from GBS data.…”

“…AFLAP analysis of a RIL population of Lactuca spp. using GBS data AFLAP was run on 235 F6 RILs generated by crossing Armenian999 ( L. serriola ) and PI251246 ( L. sativa ), which had previously been sequenced using GBS with 249.8 Mb generated for each line (12). It was not possible to obtain the single copy boundaries from GBS data as had been calculated for A. thaliana .…”

“…GBS/RADseq can be used to obtain high coverage, reduced representation sequencing data, and using ustacks, linkage analysis may be performed without an assembly (17); however, library preparation for GBS/RADseq may introduce allele bias caused by restriction site distribution (18), a lack of robust genotype calls, and much lower marker density. The utility of AFLAP was demonstrated on a lettuce RIL population genotyped using GBS (12). AFLAP generated nine linkage groups for each parent colinear with the 2.4 Gb genome assembly ( Figure 4) (13).…”

“…Low coverage WGS reads of F 2 individuals generated from Col x Ler were obtained from previous studies (9-11); 100 individuals with the largest gzipped files were selected for analysis (Additional File 2: Table S1). Reads for parents and RILs of a previously analyzed L. serriola Armenian999 x L. sativa PI251246 interspecific-cross were downloaded from NCBI BioProjects PRJNA642889, PRJNA510128, and PRJNA478460 (12).…”

“…Low coverage WGS reads of F2 individuals generated from Colx Ler were obtained from previous studies(9)(10)(11); 100 individuals with the largest gzipped files were selected for analysis (Additional File 2:Table S1). Reads for parents and RILs of a previously analyzed L. serriola Armenian999 x L. sativa PI251246 interspecific-cross were downloaded from NCBI BioProjects PRJNA642889, PRJNA510128, and PRJNA478460(12).Two B. lactucae crosses were analyzed with parental isolate SF5 in common. For both crosses, WGS of B. lactucae parental isolates SF5, C82P24, and C98O622b have been described previously(6); NCBI BioProject PRJNA387613).…”