← PortfolioThe body and the brain

Orchid

Whole-genome sequencing for IVF embryos

Team

Noor SiddiquiCEO

Founded

2020

Invested

2022

Links

Website · LinkedIn

The problem

How much can you learn about a child from five cells of an embryo?

Point at any chromosome to see both of its copies up close. Press and hold to see the small slice that standard testing reads instead.About five cells are taken from the outer layer of one embryo and their DNA is copied over and over, unevenly. Standard testing reads a small slice of it. Orchid aims to read nearly the whole genome from the same cells and put chromosome counts, single-gene conditions and polygenic scores in one report, and up close you can see where one copy of a gene failed to copy.An illustration, not real data.

How embryo testing works

IVF starts with hormones that coax the ovaries into making several eggs at once. The eggs are collected straight from the ovaries under ultrasound guidance, mixed with sperm, and the embryos that result are grown in the lab. Many are kept until the stage, five or six days in.

If the embryos are to be tested, an embryologist takes a few cells from each. Taking them at the blastocyst stage is gentler on the embryo than earlier biopsies, and the cells come from the , the outer layer that goes on to form the placenta. The usual sample is about five cells. The most common test counts chromosomes, looking for : an extra or missing chromosome. Another looks for one specific gene variant that a family already knows it carries.

Genetic disease comes in two broad flavours. A condition is caused by a mutation in a single gene. A polygenic one comes from many genes at once, and polygenic conditions are the more common kind. For those, the best genetics can offer is a : many small variants across the genome added up into a single number.

Further reading In vitro fertilisation (Wikipedia)Preimplantation genetic diagnosis (Wikipedia)Whole Genome Embryo Screening for IVF (Orchid)Genetic disorder (Wikipedia)Use of preimplantation genetic testing for polygenic disorders (PGT-P): an Ethics Committee opinion (American Society for Reproductive Medicine)The Science Behind Whole Genome Embryo Sequencing (Orchid)

Why it is hard

i.
Photocopying a whisper
Five cells hold very little DNA, far too little to sequence directly, so labs copy it millions of times first. One common method can rapidly amplify minute amounts of DNA, but the copying is uneven. Sometimes one of the two copies of a gene simply fails to amplify, a problem called , and studies of one popular method have reported dropout rates anywhere from 0 to 60%. If the copy that drops out is the one carrying a mutation, an embryo can be read as healthy or affected depending on which copy made it through.
ii.
The sample may not match
The biopsy comes from the placenta side, and the baby comes from the inner cell mass. The two can differ. By one meta-analysis of more than 800 embryos, about three-quarters of early embryos are , meaning their cells do not all share the same chromosomes. Healthy babies have been born from embryos that testing had labelled aneuploid.
iii.
Siblings are a lot alike
Embryos from the same two parents are siblings, and siblings vary less than strangers. That caps how much choosing between them can change the odds. The chance that a couple ends up with one embryo in the top fifth of scores and another in the bottom fifth has been put at under 3%. The scores also know nothing about the environment a child will grow up in.
iv.
Scores that travel poorly
Most of the big studies that polygenic scores are built from were done in people of European ancestry. How well the scores work for everyone else is unknown, which is a scientific limit and a fairness problem at once.

Further reading Multiple displacement amplification (Wikipedia)Preimplantation genetic diagnosis (Wikipedia)Polygenic embryo screening: quo vadis? (Journal of Assisted Reproduction and Genetics (via PubMed Central))

What Orchid is after

Standard embryo testing reads a small slice of the genome. Orchid's own figure is that it screens "0.25% of an embryo's genome", and only in rare cases looks for a single monogenic condition the couple is known to carry.

Orchid wants to read nearly the whole genome from the same five-cell biopsy, and put chromosome counts, single-gene conditions and polygenic scores into one report. It is clear that this has limits: no genetic test can guarantee a healthy child, and embryo screening does not replace prenatal testing.

Further reading The Science Behind Whole Genome Embryo Sequencing (Orchid)Whole Genome Embryo Screening for IVF (Orchid)

How they go at it

Step 1: Same biopsy, whole genome
The lab works from the roughly five trophectoderm cells a clinic would send for standard testing, and aims to read over 99% of the embryo's DNA. That coverage figure comes from validation studies on reference samples from the US National Institute of Standards and Technology.
Step 2: A long list of typos
The report checks more than 1,700 genes linked to severe single-gene conditions, including neurodevelopmental disorders, hereditary cancers and birth defects. Orchid's own estimates are a useful reality check: a single genetic cause accounts for only about 30% of neurodevelopmental disorders, so most cases would not show up this way.
Step 3: Scores, with a counsellor
The report also carries polygenic risk scores for conditions such as type 2 diabetes, schizophrenia and breast cancer. Parents meet a board-certified genetic counsellor before testing and again when the results come back.

Further reading Whole Genome Embryo Screening for IVF (Orchid)The Science Behind Whole Genome Embryo Sequencing (Orchid)

Still open

Does choosing embryos by polygenic score lower disease in the children born?
Several published studies suggest some relative risk reduction is possible. But the major professional bodies, including the American College of Medical Genetics, the European Society of Human Genetics and the European Society of Human Reproduction and Embryology, have called clinical use premature, and the American Society for Reproductive Medicine calls it nascent and unproven. Settling it would take long-term studies that follow embryos into adult health.
What should parents do with a probability?
Scores can be used to rank embryos or to discard some, and the two lead to different conversations. The embryo with the best score may not lead to a pregnancy, so a less favoured one may be transferred anyway. Beyond the clinic sit harder ethical worries that societies are still arguing through.

Further reading Polygenic embryo screening: quo vadis? (Journal of Assisted Reproduction and Genetics (via PubMed Central))Use of preimplantation genetic testing for polygenic disorders (PGT-P): an Ethics Committee opinion (American Society for Reproductive Medicine)

About Orchid

Orchid Health is building whole-genome sequencing (WGS) for IVF embryos, fundamentally expanding what parents and clinicians can understand about genetic risk before implantation. By sequencing the full genome rather than screening for a limited set of conditions, Orchid provides a step-change in the amount and quality of information available during family planning.

The company integrates directly into fertility clinic workflows, combining clinical genetics, data infrastructure, and AI-driven interpretation to deliver results that are actionable, compliant, and practical to use. Orchid is not positioning genomics as research — it is making it part of standard care.

Orchid is led by Noor Siddiqui, a founder with rare conviction in a sensitive and consequential domain. The company has been covered by WIRED and The Washington Post and is working with fertility clinics across the country.

Words used here

blastocyst: An embryo about five or six days after fertilisation, with an outer layer of cells and an inner clump that becomes the baby.
trophectoderm: The outer layer of cells in a blastocyst, which goes on to form the placenta.
aneuploidy: Having an extra or missing chromosome.
monogenic: Caused by a change in a single gene.
polygenic score: A number that adds up the small effects of many genetic variants to estimate a person's inherited risk of a condition.
allelic dropout: When one of a gene's two copies fails to copy during DNA amplification, so the test only sees the other.
mosaic: Made of cells that do not all share the same genetic makeup.

Sources

How much can you learn about a child from five cells of an embryo?

Photocopying a whisper

The sample may not match

Siblings are a lot alike

Scores that travel poorly