Pawpaw breeding considerations

General Fruit Growing
#21

Not having read the paper how did they decide on the genetic markers to select?

#22

@Jujube
H. Huang conducted 4 years of research into RAPD markers for Pawpaw: RAPD Inheritance and Diversity in Pawpaw (2000). The testing was performed in the following year: Molecular Characterization of Cultivated Pawpaw (2003).

Pomper et al did zero research. Instead they contracted with Genetic Identification Services of Chatsworth CA for selection of markers and production of marker data.

#23

Interesting but I personally wouldn’t be so quick to dismiss a lab for contracting out some experimentation.

Regardless this is interesting for ancestry purposes but phenotypic relation overlayed on this would really boost this researches reach

1 Like
#24

They did not know enough to reject the results.

Unfortunately the lab used by H. Huang only returned 49 of his markers free from error. I have re-analyzed that data and produced what I believe is a bias-free set of 45 markers - to be published by IJCSA next month.

Next I plan to apply all of H. Huang’s markers to the 100 available cultivars in circulation and incorporate morphology as well. At current pricing this would run about $40k for error-free results.

2 Likes
#25

Here is another look at Sweet Alice. Values from 2 to 7 are considered close, from 8 to 11 are moderately close, 12 to 13 are sub-average. Keep in mind the measurements are coarse - so they might only be accurate to +/- 2.

Of interest here is the isolation of Sweet Alice. Only Overleese and Sunflower can be considered independent finds and the remainder are due to ancestral parentage.

cultivars of interest - Sweet Alice
cultivars of interest - Sweet Alice1500×1347 117 KB

#26

Again interesting for ancestry purposes but how the markers correlate to phenotype would be even more interesting. Additionally the trends in which markers are shared or not may be informative if patterns emerge.

Additionally, I suggest adjusting the node distances to be proportional to the number of mismatches (more fibonacci looking) , make lines gray, and bold “Sweet Alice” to improve figure readability. Just a suggestion

1 Like
#27

I agree. That is the purpose of this study.

It is a topological graph extracted from a 45-dimensional space. Note that the displayed rotational sequence of neighbors is arbitrary and likely has no relation to the orientation of any 2D projection.

I agree.

Asimina (pawpaw) and Annona pollination biology references
#28

I’ve now worked out the primary ancestry group for each of the cultivars tested by H. Huang. Here’s a short table of how they compare with the questionable clades published by KSU.

[Updated 6/2/2022]

Cultivar Year introduced Primary association KSU clade
Overleese 1950 A V
NC-1 1976 A V
Prolific 1985 A II
Rappahannock 1990 A III
Shenandoah 1990 A V
Susquehanna 1990 A II
Potomac 1994 A III
Sweet Alice 1945 B III
Taylor 1968 C I
Taytwo 1968 C V
Sunflower 1970 D V
8-20 1994 D II
Wabash 1994 D III
Rebecca’s Gold 1974 E V
Mitchell 1979 F V
PA-Golden 1986 F untested
Wells 1990 F IV
3 Likes
#29

Here’s a list of cultivars I believe should be in any pawpaw breeding program, plus any advanced specimens you wish to fine-tune:

Middletown
Sweet Alice
Taylor
Sunflower
Rebecca’s Gold
Mitchell

1 Like
#30

As a general observation only, I think some of those are hard to find readily as an end consumer.
They exist perhaps mainly at repositories and collectors’ orchards.
Though as an academic/research endeavor this would probably not matter.

#31

Each were offered at some point in the last 12 months at one or more retail nurseries.

Frost, R. Diversity of Pawpaw (Asimina triloba) cultivars in USDA repositories and selected retail nurseries c. 2022 (in preparation).

1 Like
#32

Richard, what is the reasoning behind your suggestion for including these varieties? Is it primarily based on genetic distance between each of them to improve the gene pool?

To me, the idea of searching for a new cultivar focuses on traits similar to what Neal Peterson was seeking- high flesh to seed ratio, low seed count, etc.

I would also consider freestone, the trait in Susquehanna limiting seeds full formation, and Allegheny’s high productivity to be important along with selecting for low acetogenins. Obviously there is a laundry list of other traits that could be selected for breeding but some are more important to different growers.

#33

Yes.

#34

For example, understanding what constitutes an induction fallacy.

#35

Accepted for publication, January 2023.

4 Likes