EIC Accelerator Winner Amatera and the Crop Breeding Race

Creating a new coffee or wine grape variety with conventional breeding can take more than 20 years and millions of euros, according to Amatera's chief executive. Amatera Biosciences, founded in 2022 by Omar Dekkiche and Lucie Kriegshauser, proposes to move that sorting step from the greenhouse to the single cell: isolate plant cells one by one, check each for the target genetic change, and regenerate only the promising ones into plants. The company calls the platform CellTera, and its website describes it as a combination of cell culture, high-throughput automation and machine learning.

CellTera appears on the EIC Accelerator selection list published on 15 June 2026, recommended for blended finance. The list is headed "EIC Accelerator 2026, 1st and 2nd Batch"; the source table used for this series dates the round 6 May 2026. CORDIS records a grant agreement signed on 28 September 2026, with an EU contribution of €2,499,962.50 for a project running from 1 October 2026 to 30 September 2028. The project is a month old, so this article sets out what it promised and what can already be checked.

Screening the cell before growing the plant

Gene editing, mutation breeding and doubled haploids all depend on regenerating plants from cultured cells, and the company's seed release lists all three as breeding routes its automation serves. The CellTera objective says breeders must regenerate and screen thousands of plants over months or years to find a few with the target trait, because current methods cannot reliably screen at the cell stage or confirm that a regenerated plant comes from one cell. Its keywords name the problem that follows: chimerism, a plant carrying a change in some tissues and not in others. CORDIS says CellTera "guarantees that regenerated plants originate from a single cell" and enables "chimerism-free, cell stage screening".

In an interview published by AgFunderNews on 28 October 2026, Dekkiche said the company has patented a pipeline to microculture single plant cells, each grown separately, and that only cell lines of interest are regenerated into plants. He was also clear about the limit: a trait can be screened at cell level only when its genetic marker is known. Caffeine production can be, because cells carry the metabolism; drought tolerance and yield cannot, and "we don't promise that". AgFunder, the publisher's parent company, is an Amatera shareholder, which the outlet discloses. A search of public sources on 4 November 2026 found no published patent under the applicant name Amatera, so the patent rests on the company's statement.

Amatera started with its own varieties. Its website lists "Robustica", a Robusta-derived coffee meant to taste closer to Arabica, a naturally caffeine-free Arabica, and a mildew-resistant Merlot. In 2023 its method induced genetic variation in cultured cells by physical and chemical means rather than gene editing; the 2026 interview says the same screening can be applied to edited cell lines.

One platform, several multipliers

The performance figures change from document to document. Paulig's October 2023 release said the technology speeds up coffee breeding four to five times. The company's March 2026 seed announcement and its press release say "up to 2x faster and 10x more cost-effective". In October 2026 Dekkiche told AgFunderNews he expects at least five times the cost saving and two times the time saving, and that lab data is still being generated. The CORDIS objective states that "CellTera accelerates trait implementation timelines by over 4x, reduces costs by 10x and increases throughput by 30x." The figures do not all measure the same thing, and none of them has been published as a measured result.

The 2023 timeline is from AgFunderNews and the three-to-five-year launch window from Paulig. The March 2026 Global AgInvesting report on the seed round likewise names no partner and says the company expects to sign several partnerships in annual crops over the following year. The same report describes a two-part business model: licensing its own perennial varieties to growers and industry partners, and acting as a technology provider to seed companies in annual crops.

Equity first, then the EIC

The founders met through the Entrepreneur First programme, according to AgFunderNews. A €1.5 million pre-seed round in October 2023 was led by PINC, the venture arm of Paulig. On 4 March 2026 the company announced €6 million in equity, co-led by Demea Sustainable Investment and Oyster Bay Venture Capital with the earlier investors. The release says Demea invested through VitiRev Innovation, a €70 million fund for the environmental transition of the wine industry.

CORDIS records the grant but gives no total project cost, and no amount for any equity component of the blended award has been published. A search on 4 November 2026 found no EIC Fund investment in Amatera.

The BODACC legal notices show accounts filed for 2022, 2023 and 2025, the last deposit published on 31 August 2026, a capital change in August 2023, the move of the registered office to Évry-Courcouronnes from 4 December 2025, and a notice of a change in administration in August 2026. No notice for the 2024 accounts appears, and no insolvency proceedings are recorded. The national company directory lists the company as active with 10 to 19 employees for 2024 and publishes no financial figures.

What CellTera promises

The EIC project is a development and demonstration programme for the platform itself, not for a coffee or grape variety: CORDIS names no crop and refers to annual and perennial breeding pipelines. Its goal is regeneration that is scalable and compatible with automation. In the AgFunderNews interview, Dekkiche described the aim as one end-to-end device that runs plant cell regeneration workflows autonomously.

1. Single-cell, chimerism-free regenerationToo earlyDescribed in the CORDIS objective; no results or publications listed yet.
2. Over 4x faster, 10x cheaper, 30x throughputClaimProject targets; the company's own releases give different multipliers.
3. Automated end-to-end deviceToo earlyStated as an aim by the CEO in October 2026.
4. Seed company partnershipsNot namedExpected "in the coming months", per the CEO in October 2026.

The European context

CORDIS tags the project at 40% for biodiversity. The wider frame is the EU's new law on plants from new genomic techniques: the Council adopted it in spring 2026, as the Commission's IP Helpdesk reported on 5 May, and the European Parliament gave its final approval on 17 June 2026, reported by ANSA. Plants in its first category are treated like conventionally bred varieties, and the rules apply two years after entry into force. CellTera's CORDIS keywords include gene editing, the technique the law covers. Amatera's own coffee and grape lines are presented as non-GMO; in 2023 the company said it induced variation by physical and chemical means rather than gene editing.

Amatera has not been asked for comment for this article. The reporting is based on publicly available sources, listed below.