Using PACBio Long-Reads to Generate High Quality References for the Allotetraploid Coffea arabica and Its Diploid Maternal Ancestor C. eugenioides: characterization of Genomic Regions Containing QTLs for Yield, Plant Height, and Bean Size

We are celebrating this year our 10^th ICGN Coffee Genomics Workshop at PAG. Over the past decade our coffee genomics community has focused efforts on bringing coffee to the forefront of plant genomics research.

The first coffee genome assembly published (Denoeud et al. 2014) was for the diploid cultivated species Coffea canephora. In parallel, Cornell University and FNC/CENICAFE, submitted a proposal to IDB/FONTAGRO to sequence the genomes of the most widely cultivated coffee species, the allotetraploid Coffea arabica, and its diploid maternal ancestor C. eugenioides. We received funding in 2016 from NSF to strengthen this effort and generate state of the art high quality reference genomes for these Coffea species to help accelerate linkage of structural and functional diversity in coffee for climate change adaptation. We used high coverage PACBio long reads for de novo genome assembly, and PCR free Illumina paired-end sequencing data for error correction using the MaSuRCA genome assembler (Zimin et al. 2013) prior to genome annotation and high through put diversity variant calling. We built de novo transcriptome assemblies for C. arabica using RNA reads of a variety of functional genomics experiments, and used these to progressively validate completeness and quality of our genome assemblies.

In addition, progress in the characterization of genomic regions containing QTLs for yield, plant height, and bean size in C. arabica through the integration of the linkage groups harboring the QTLs (Moncada et al. 2016 Tree Genetics and Genomes 12: 5 DOI 10.1007/s11295-015-0927-1), the physical map of the species (contracted by CENICAFE to Rod Wing at U. of Arizona), genomic sequences obtained by whole genome and targeted sequencing (BAC by BAC), and full length transcriptome (Iso-Seq method, PACBio) from the progenitors of the population used to detect the QTLs will be presented. This analysis will provide candidate genes and genomic features related to important agronomic traits in the allotetraploid C. arabica, useful for functional genomics and to develop tools for marker assisted selection.

The PACBio coffee genome assemblies were done in collaboration with Pacific BioSciences and the Colombian Center for Bioinfomatics and Computational Biology (Bios).

This abstract will have an extended time (40 min) and will be presented by co-authors M. Yepes (project introduction PACBio assemblies), A. Zimin (hybrid genome assemblies error correction), K. Mockaitis (transcriptome assemblies), and C. Maldonado (characterization genomic regions containing QTLs for traits of interest).