Reading the Whole Letter

October 2026

Researched and drafted with AI assistance, then reviewed by me. ← Back to Knowledge Base

Every child gets a long letter from each parent. The letter is written in DNA. It has about three billion letters, and it tells the body how to build itself. Most of the time the letter is fine. Sometimes one letter is wrong, and a child is born with a disease nobody saw coming.

Here is the odd thing. Reading an adult’s whole letter has been routine for years. But for an embryo, the point where the letter matters most, we mostly read a few lines.

Couples who do IVF often make several embryos. The clinic must pick one to transfer. How do they pick? Mostly by looking at the embryos under a microscope and by a test that counts chromosomes. That test reads well under one percent of the genome. It is like checking that a book has all its chapters without reading any of the pages.

That is the problem Orchid, the company Noor Siddiqui founded, works on. Orchid takes about five cells from an embryo and reads almost the whole genome. It checks for more than 1,200 diseases caused by a single broken gene. It also gives risk scores for common diseases, like some cancers or diabetes, that come from many genes acting together.

Her reason is personal. Her mother has an inherited eye disease that slowly takes away sight. Many parents of a child with a rare genetic disease had no idea they carried it. You cannot avoid a risk you don’t know about.

Why didn’t this exist before? Two reasons. Reading DNA used to be expensive. And five cells is a tiny amount of DNA. You must copy it many times before you can read it, and copying makes mistakes. Orchid’s bet is that this can now be done well enough to be useful. Today it costs $2,500 per embryo, on top of IVF that already costs tens of thousands. About 100 US clinics offer it.

The case for it is simple, and the podcast makes it to both political sides. If you believe “my body, my choice,” this is a choice. If you believe the government should stay out of private life, this is private life. Parents already pick embryos. The only question is whether they pick with more information or less.

But the critics are serious, and their worry is not only ethics. Some geneticists say copying DNA from five cells adds errors, so a healthy embryo might look sick, or the reverse. The bigger doubt is about the scores for common diseases. These scores describe groups well but predict one person poorly. A 2024 study found that different scoring methods often picked different “best” embryos. The scores also work less well for people without European ancestry, because most of the data comes from Europeans. The main US medical genetics body says the benefit is unproven. Some people see the whole idea as a step toward eugenics, which Orchid rejects.

I think the fair way to see it is this. Finding one broken gene that causes a serious disease is a real, narrow good. Ranking embryos by their chance of a common disease is a much weaker tool, and it is easy to sell it as more than it is. The danger isn’t that parents get information. It is that they get numbers that look more certain than they are.

Still, notice the direction. Every tool that reads something starts out expensive, noisy and controversial, then becomes cheap and boring. If that happens here, the question won’t be whether parents can read the whole letter. It will be what they choose to do with what they read.

Sources: Freethink: Screening IVF embryos for 1000+ diseases · CNBC: Orchid launches embryo genome test · Washington Post: Inside the Silicon Valley push to breed super-babies · Scimex: Genetic risk scores for embryo selection · Exaudi: Scientific criticism of embryo selection · The Network State Podcast #35