As a fellow scientist, I have to agree that science can do (1), (2), and (4) — because we have many examples of each.
But with (3), you are seemingly confusing results which indicate that we are incredibly similar genetically, with selecting appropriate social responses to pair-bonding and mating behavior.
This is unwarranted. I happen to agree with your point of view, but it’s *not* the result of science, not in the form you state. Science tells us that races *are a thing* inasmuch as you can determine someone’s origins via SNP maps (but “races” defined that way don’t perfectly align with the colloquial notion). It *also* tells us that men and women *are different* in countless ways. It *also* tells us that heterosexuality, not homosexuality, is typical. That doesn’t mean that racism is okay, that sexism is okay, that homophobia is okay. To conclude as much would be to commit the naturalistic fallacy — but you’re doing the same in the opposite direction.
Let’s just not do that. This kind of replacing of data with emotion undermines the scientific enterprise. There’s no shame in being tolerant because we believe that this leads to a better society even if we don’t have extensive control and experimental studies where we run societies different ways. If you want to say, “Science shows us that we are incredibly similar genetically and in other ways,” go for it! If you want to draw the inference that demonizing others as “inhuman” is factually incorrect, go for it.
And if you want to promote a just and tolerant socienty, go for it. Just don’t mislead yourself and others into thinking that we have solid evidence that this is the factually correct way and other ways are wrong.