His audience is roughly an even split between people who are incredibly skeptical and people who don't have enough context to be skeptical.

(Anyone who'd be unimpressed by this feat has other things to do with their time – this is Introduction to Genomic Engineering, not Advanced Topics in Genomic Engineering.)

"The important regions of DNA have shockingly little variation. The building blocks of life are time-tested, rarely needing improvement: nuclei, ribosomes, mitochondria, vacuoles, and so on. The disparity of species – and individuals, for that matter – comes from a relatively small number of what are called morphogenic proteins. Like drops of blood in a glass of water, morphogens create overlapping chemical gradients that cells can read to learn their location and identity. The rest of the body plan is arranged by cells communicating with their neighbors, activating some systems and silencing others, selectively bonding to form membranes and mechanical parts. It is remarkably elegant.

"Learning what morphogens do is a matter of deleting them from the genome and observing what does not happen. If you are the first to try this for any given protein, please refrain from immediately naming it after the first thing you see it do, as morphogens are each responsible for several aspects of development. The morphogen system that handles the dorsal-to-ventral axis was initially called wingless, because of how fruit flies grew when it was damaged. It was later discovered to be responsible for far more than that and was renamed wingless-related integration, because biologists from my homeworld do not learn from their mistakes.

"Once you have determined all the morphogenic systems involved in the patterning of your choice, it is not enough to splice those genes wholesale into a new host. Activating those genes is contextual – their promoters are tuned for cell locations prescribed by a higher level of morphogenic direction in the original species. When you are more comfortable with genomic engineering you will be able to design new binding domains that operate the donor genes in tune with the host's body plan, but as a beginner I found it much easier to transfect cells as the embryo developed."

Next slide.

That… is either a picture of a human infant with an elephantine trunk, or a convincing forgery.

Morgan has several questions, two of which are "why" and "how". She is admittedly also curious as to what complications the procedure had, and whether this man's homeworld is so infested with abnormal human physiology that he felt he could get away with this in the long term. Is this what happens on Earths without normalcy organizations?

"An elephant's trunk is a muscular hydrostat, much like a human's tongue. They call for some clever design tradeoffs, which have not only been solved by evolution but implemented in both species. By transfecting the morphogens for muscular hydrostat development into the cells of the frontonasal prominence, the region began to develop more like a tongue than a nose. Additional morphogens were required to ensure the nasal passages formed in the body of the trunk, but at the level of muscle groups it was unnecessary to micromanage the transition. Paired muscles were able to identify one another and grow in tandem using existing human gene regulatory networks – a shortcut which cannot always be counted on, especially for more exotic modifications – and so on for innervation, angiogenesis, and the other necessary features of any body part.

"Bear in mind that this will take trial and error. Finding the morphogens you need is the easy part. The hard part is using them without damaging other critical systems. It does you no good to add the genes you want if the resulting organism is born with four heart defects and no kidneys. Less is more."

What is with these people and wanton human experimentation? First Dr. West, then Riley, now this armored megalomaniac. Couldn't he have made a dog with an elephant's trunk instead?

"You waited until week four to turn a baby into a GMO? What about placentation, how did you reimplant it without terminating?"

"The procedure was conducted in utero."

"Whoa. Where'd you find the mother who agreed to that?"

"I ran an advertisement in the classifieds of the New York Times," he says, irritated by the repeat interruption. Finding a woman willing to put up with an abnormal pregnancy For Science™ was not the onerous part – arranging for childcare was much, much harder. "It took less than—"

"The nose is much shorter than a young elephant's nose, if they were the same size," Semirhage says thoughtfully. "The width also, it is not the same."

"Elephants are much larger than humans; there was far less antenatal hormone exposure, even accounting for the mother's exogenous sex steroids. There may also have been—"

"How did you get enough of the cells expressing elephant growth factor to stimulate the process in the first place?"

"Did the primary palate form correctly?"

"Why are the eyes—"

He taps the microphone twice, sparking a painfully loud burst of feedback.

"Thank you," he says, once everyone's taken the hint. "There were no issues with the primary palate. One instance of maxillofacial surgery was necessary to repair cosmetic defects, but otherwise the experiment was a success. As to the implementation details…"

"… in this case we used a modified adeno-associated virus, which has the useful property of not integrating the elephant DNA into any host chromosomes. Wild type AAV can integrate, targeting 19q13, but for embryonic modifications such integration is unwanted. Localized expression of the elephant morphogens during the pregnancy itself was sufficient – the trunk's development post-partum was accomplished entirely via human proteins. AAV has a limited payload size, which can be problematic if a single segment of foreign DNA is more than 4,400 bases in length, but this will usually not be the case. The actual injection was several different AAVs in the same micropipette."

"There are more ways to tinker with DNA than you have months in which to learn them, but for our purposes there is a single best method. Transforming embryos is highly specialized, done by clamping a single cell in place and piercing it with a micrometer pipette…"

What follows is a remarkably ordinary explanation of how to use IVF tools to transfect mammal cells without heat shocking them first, by way of using a very small needle to pierce the nucleus. Once inside the nucleus, reverse transcriptases in the payload incorporate the foreign DNA into the genome, trading increased precision for the need to modify each embryo by hand. It's a worthwhile exchange, if you know what you're doing and plan accordingly.

This feels like being back in embryology class, only this time her instructor is a passionate genius instead of a geriatric moron with too many malpractice lawsuits under his belt. She pays rapt attention, typing notes into her phone as fast as she can.

The lecture concludes with a few remarks on existing research for aspiring newcomers to the field, followed by the inevitable.

"Does anyone have any questions?"

Most people raise their hands. Elliot just starts talking.

"You've implied that the rate of off-target edits is a rounding error – definitely not my experience with germline cas9 editing, we had fatal mutations about half the time, made it impossible to do more than two or three genes per embryo. What was the solution?"

"This was an issue with earlier versions of the CRISPR method using the original enzyme from S. thermophilus. Modified versions of cas9 with lower DNA binding affinity have much lower off-target rates, closer to ten percent for each gene, and restriction endonucleases specific to the desired target site have virtually no problems whatsoever. We still had some failures, mostly due to equipment malfunction, but customizing everything beforehand substituted a lot of paperwork for even more labwork."

"Where were you ordering the enzymes from?"

"Oscorp and Illumina-Fisher. Anyone else?"

"Did you use the same process for all the uplifts or is it unique to each species?"

He emits a computer-modulated sigh.

"Everyone who has questions about uplifts, please lower your hands."

Almost everyone lowers their hands.

Typical.

"True intelligence enhancement is its own subject. I will request one of tomorrow's open seminar slots to give it a more thorough treatment, since there appears to be some demand for it. Does anyone have any questions unrelated to uplifts?"

"How did you discover so many safe harbor sites? What was the payload division scheme, by chromosome or by upstream dependence? Did you grow all of the embryos in vivo, or just the human ones? Was this research legal? Where was your revenue coming from?"