Note: this post is still a rough draft but I think this idea is important enough to publish it anyway. I intend to revise it in the future. Also if you are a biologist who wants to collaborate with me on this project, or a philanthropist who wants to provide funding, please email me: metacelsus at protonmail dot com
Gwern Branwen, Carl Shulman, and others have written about the concept of iterated embryo selection (IES). The idea of IES is to:
Take some embryos,
derive embryonic stem cell lines from them,
genotype the cells and select the best ones according to a polygenic score,
differentiate the stem cells to gametes,
generate a new population of embryos and repeat.
IES would allow a much more rapid improvement in genotype than would be possible under normal selective breeding.
But there may be a faster and better way to do this (and it might even be easier)!
The idea: iterated meiotic selection
Here is my proposal:
Take a diploid cell line (probably ESC or iPSC or PGCLC)
Induce meiosis and form many haploid cell lines.1
Genotype the haploid lines and select the best ones.
Fuse two haploid cells to re-generate a diploid cell line.
Repeat as desired. At the end, either differentiate the cells into oocytes or perform nuclear transfer into a donor oocyte.
This would have several advantages.
Selection is more powerful at the haploid stage.
Meiosis is faster than gametogenesis.
This does not require in vitro gametogenesis technology, since the final generation of an embryo could be performed by somatic cell nuclear transfer into a donor oocyte.
In addition to enabling iterated selection, this would also allow “genetic crosses” for human cells. Genetic crosses are very useful for model organisms (e.g. crossing Cre/lox mouse lines to generate tissue-specific knockouts) and they would be similarly useful for human cell lines.
Technological challenges and prospects
The key piece of missing technology is a way to induce meiosis in a diploid cell line to generate haploid lines. Other pieces (such as culturing haploid human stem cells, or electrofusion of cells) are already in place. Therefore, meiosis is all you need.
Now, inducing meiosis won’t be easy. Meiosis normally takes place only during gametogenesis and it is unclear whether it is possible to activate meiosis separately. But I think that it has a good chance of being achievable, and might even be easier than in vitro gametogenesis (because it’s easier to screen for a haploid cell than for a viable gamete).
I will approach this problem by:
Performing a literature search and computational analysis to identify proteins involved in meiosis.
Constructing a barcoded transposon library for inducible expression of these proteins.
Integrating the library into iPSC lines with reporters for meiosis-specific proteins (e.g. REC8)2
Inducing expression and sorting reporter-positive cells. Haploid cells could also be sorted by staining for DNA content, although this has a smaller dynamic range than fluorescent reporters.
Sequencing barcodes to identify the top factors.
Validating the top factors by expressing them individually or in small combinations.
This overall approach has worked before to generate a variety of cell types (including granulosa cells and oogonia)3 and I think it has a decent chance of success on this problem.4
Note that these would have to be female (X-bearing) since the X chromosome contains essential genes.
REC8 is a meiosis-specific cohesin. I actually already made a REC8 reporter line as part of my gametogenesis project.
Preprints to be released shortly.
It may not be feasible to induce meiosis directly from iPSCs. However, I think that this has a high chance of working in iPSC-derived oogonia or spermatogonia, although these are more difficult to culture.
Going through multiple generations without development of a full organism, selecting only on what are thought to be good traits in the genotype, seems like it would be vulnerable to mutations (or just combinations of existing alleles) that make the full organism non-viable, or weak.
For "lower" organisms, I guess one could try it and see how much of a problem this is, but for humans it seems like this is an ethical problem with this technique (hardly the only one, of course).
Any chance you want to do a slightly more in-depth write-up to guest author on my substack?