# An ancient gene times avocado flowers

> A genomic study links avocado's complementary A- and B-type flower schedules to two SDMYB haplotypes maintained for more than 42 million years, while stopping short of direct gene-editing proof.

_Source: Peer-reviewed PNAS study, verified against the full text in PubMed Central, PubMed, Crossref and OpenAlex · 2026-09-04 · 7 min read · Verified against primary sources_

Canonical: https://iyu.app/e/ancient-gene-times-avocado-flowers-2026

## The 60-second version

Genome mapping and time-series expression data link avocado's A/B flower schedules to two ancient SDMYB haplotypes.

**Key points**

- The primary association analysis used 374 resequenced offspring and found one strong locus on chromosome 10.
- The dominant allele's expression rhythm lagged by about two hours, matching the delayed second flower opening in A-type trees.
- Comparative genomics estimates the haplotypes have persisted for more than 42 million years and occur in at least 26 non-avocado species.
- Markers may classify seedlings before the crop's five-to-twelve-year juvenile phase ends, but yield gains have not been demonstrated.

**Verdict.** SDMYB is the strongest supported regulator and a useful marker candidate, while direct functional conversion and field benefit remain to be tested.

## Full explainer

A PNAS study links avocado's complementary A- and B-type flowering schedules to two ancient haplotypes around **SDMYB**, a flower-specific transcription factor. The evidence could let breeders identify flowering type in seedlings, but it does not yet show that changing this gene alone rewires the schedule or raises orchard yields.


### The flower clock — One flower, two phases

An avocado flower is hermaphroditic. It first opens in a female phase, closes, and reopens the next day in a male phase. Under typical conditions, A-type trees begin the female phase in the morning; B-type trees follow the complementary schedule.

> **i** The labels describe flowering schedules, not separate female and male trees. Temperature and light can shift the timing, so real orchards do not always match a rigid clock.


### How it was found — A signal on chromosome 10

The researchers ran a genome-wide association study in 374 resequenced offspring from an open-pollinated cross between A-type Gem and B-type Luna UCR trees. One strong association appeared on chromosome 10. Read-depth patterns in a broader group of 504 phenotyped and resequenced relatives supported a structural difference at the same locus.

- **374** — offspring in the primary genome-wide association analysis
- **~2 h** — average expression-phase delay of the dominant A1 allele
- **>42 My** — estimated persistence of the two haplotype lineages
- **26+** — non-avocado species carrying diagnostic variants

Additional avocado genomes narrowed the associated interval to one leading candidate: **SDMYB**, a member of a transcription-factor family involved in late flower maturation. A separate mapping panel published while this study was under review independently pointed to the same gene region.


### Mechanism — Expression moves with the schedule

The team sampled flowers from three biological replicates of each type about every three hours across a daily cycle. SDMYB expression was rhythmic in both types. In A-type trees, the dominant A1 allele's expression rhythm lagged the A2 allele by about two hours, close to the delay in the flowers' second opening.

- **Mapping:** A single chromosome 10 region tracked A- versus B-type flowering in the primary family.
- **Structure:** Read depth and assembled genomes showed distinct haplotypes around the locus.
- **Expression:** SDMYB was flower-specific and its two alleles followed different daily phases.
- **Function:** No reported knockout or allele-swap experiment directly converted one flowering schedule into the other.

> The result is a chain of converging evidence, not a one-experiment proof of a solitary genetic switch.


### Evolution — A polymorphism older than the crop

Comparative genomes and a fossil-calibrated phylogeny place the divergence of the two haplotype lineages more than 42 million years ago. Diagnostic variants occur in at least 26 non-avocado species in the Perseeae tribe.

The proposed stabilizer is negative frequency-dependent selection. A rare flowering type can gain more compatible pollen partners, preventing either schedule from taking over. The 42-million-year figure is a molecular-clock estimate with model and calibration assumptions, not a direct measurement.


### What changes — Seedlings could be screened early

Avocado's juvenile phase lasts roughly five to twelve years. Genetic markers at the SD locus could identify A- or B-type seedlings before they flower, reducing the time and land needed to evaluate crosses and choose pollen donors.

> **i** The paper proposes a breeding tool; it does not report a field trial in which marker selection increased fruit yield. Local climate still affects when flowers open and whether A/B overlap is useful.


### Bottom line — Use the marker, test the outcome

Breeders can now test the locus as an early screening marker instead of waiting years for every seedling to flower. The next evidence should show whether that shortcut improves crossing efficiency and orchard performance across climates.


## Primary sources

- [Telegram post 1463](https://t.me/CNSmydream/1463)
- [PNAS paper (DOI 10.1073/pnas.2606876123)](https://doi.org/10.1073/pnas.2606876123)
- [PubMed Central full text](https://pmc.ncbi.nlm.nih.gov/articles/PMC13438604/)
- [PubMed record (PMID 42520124)](https://pubmed.ncbi.nlm.nih.gov/42520124/)
- [Crossref metadata](https://api.crossref.org/works/10.1073/pnas.2606876123)
- [OpenAlex record](https://openalex.org/W7171506333)

---
_Published by iyu (https://iyu.app) — the day's AI news, checked against primary sources and rewritten in plain language. Free to quote with attribution and a link to the canonical URL._