Studying the consequences of hybridization on plant performance is insightful to understand the adaptive potential of populations, notably at local scales. Due to reduced effective recombination, predominantly selfing species are organized in highly homozygous multi-locus-genotypes (or lines) that accumulate genetic differentiation both among- and within-populations. This high level of homozygosity facilitates the dissection of the genetic basis of hybrid performance in highly selfing species, which gives insights into the mechanisms of reproductive isolation between lines. Here, we explored the fitness consequences of hybridization events between natural inbred lines of the predominantly selfing species Medicago truncatula, at both within- and among-populations scales. We found that hybridization has opposite effects pending on studied fitness proxies, with dry mass showing heterosis, and seed production showing outbreeding depression. Although we found significant patterns of heterosis and outbreeding depression, they did not differ significantly for within- compared to among-population crosses. Family-based analyses allowed us to determine that hybrid differentiation was mostly due to dominance and epistasis. Dominance and/or dominant epistatic interactions increased dry mass, while decreasing seed production, and recessive epistatic interactions mostly had a positive effect on both fitness proxies. Our results illustrate how genetic incompatibilities can accumulate at a very local scale among multi-locus-genotypes, and how non-additive genetic effects contributes to heterosis and outbreeding depression.

This file contains information about the dataset used for the article:

"Fitness consequences of hybridization in a predominantly selfing species: insights into the role of dominance and epistatic incompatibilities"

Published in Heredity, by Josselin Clo, Joëlle Ronfort and Laurène Gay. For any questions, please contact: josselin.clo@gmail.com

The dataset is composed of two files:

-The main dataset ("data.csv")
-The flowering data ("flower_data.csv")

## For flowering ##

It contains only two columns, giving the flowering date of the 2000 individuals (same order as in the main dataset).
The first column is the number of days since 01.01.1970, and the second column is the number of days since 01.01.2018 (these values were used in the
manuscript).

## For the dataset ##

It contains 9 columns:

"female_ID" = The ID of the female plants. For consistency with the manuscript 163 = SPA, 18 = FR, 89 = CO3, 43 = CO2, 02 = CO1. Then XXX-1 and XXX-2
refers to the MLG number 1 or 2, for example.

"male_ID" = same as female, but for the pollen donor

"generation" = The generation of the cross of the individual (P= Parental line, F1, or F2)

"dry_mass" = The dry mass of the plant (in g)

"graine_tot_complet" = The number of seeds produced by a plant

"length_shoot_flowering" = the length of the main shoot at flowering (in cm, not used in the analyses)

"seed_weight" = the weight of the seeds (in g, not used in the analyses, and not very precise)

"bloc" = A number between 1 and 8 referring to the 8 blocs of the greenhouse

"zone" = A number between 1 and 2 referring to the 2 zones of the greenhouse