|
@HildeSchneemann | |||||
|
“The Geometry and Genetics of Hybridization” -- a thread: How much can we learn from patterns of hybrid fitness? Let’s use a fitness landscape models to generate some predictions and see how they depend on the process of divergence (1/n) pic.twitter.com/z2muS83l03
|
||||||
|
||||||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
2/n Strikingly, the patterns of hybrid fitness are robust to a wide range of divergence scenarios including divergence with gene flow, system drift, inbreeding, environmental change, and segregating variation. pic.twitter.com/UpvZD6Bmgb
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
3/nWe can predict fitness differences between hybrids from only a few measures of the parental phenotypes, corresponding to distances in the fitness landscape; loosely: the fitness of the parents (their distance to the optimum) and the phenotypic distance between the parents: r12 pic.twitter.com/C6EdLYVS0V
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
4/n One of these parameters, r12, tells us whether parental phenotypes have diverged more or less than would be predicted by a random walk model of evolution. Its value tends to shrink over time due to system drift or complex environmental change. pic.twitter.com/b6fXCw4goR
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
5/n High values (red) of r12 can lead to bounded hybrid advantage, heterosis, or ecological isolation. Low values (blue) lead to a characteristic pattern of intrinsic isolation. pic.twitter.com/lRrSkVFIgm
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
6/n The distance parameters in our model can be translated into QG composite effects, and estimated from hybrid cross data. pic.twitter.com/PwBCeaf6zf
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
7/n Overall, this model gives a decomposition of hybrid fitness that (i) has a clear biological interpretation, (ii) corresponds to distances in a fitness landscape model, and (iii) can be estimated from empirical data.
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
8/n Big thanks to Bianca for her persistence on rigorous maths, Nicolas for enthusiasm, inspiration and biological reality, Denis for his awesome sim. code, and to John who was patient enough to answer my endless questions, finally resulting in some fruitful ideas!
|
||
|
|
||
|
Hilde Schneemann
@HildeSchneemann
|
5. pro |
|
Link to pre-print: biorxiv.org/content/10.110… Looking forward to any comments/thoughts!
|
||
|
|
||