Synthetic Chromosomes in Yeast: A Platform for Understanding and Programming Eukaryotic Life Authors: Daniel T Lusk, Ian M Ehrenreich Published: 2026 ABSTRACT ABSTRACTA central goal of biology is to understand how genome sequence, content and organization give rise to living systems with specific traits. Achieving this goal requires understanding which genes are collectively required for life and for particular traits, how their products work together and how the organization of these genes in the genome affects cellular function. At a broader scale, it also means determining the mechanisms that generate inherited phenotypic diversity and enable adaptation and speciation. Synthetic chromosome assembly in the budding yeast Saccharomyces cerevisiae provides a tractable platform for probing these relationships in a model eukaryote by allowing chromosomes to be designed, built and tested directly in living cells. By making chromosome‐scale genome architecture an experimental variable, chromosome synthesis differs from approaches that describe organisms as they occur in nature or test the effects of one or a few genes at a time. This review traces the emergence of yeast synthetic chromosomes as tools for genome biology and identifies advances needed to make chromosome‐scale design more predictive. Introduction Chromosomes shape cellular function by defining the genes a cell carries and the genomic context in which those genes are regulated and act together (Sexton and Cavalli 2015; Coradini et al. 2020; Brooks et al. 2022). The same chromosomal context also shapes evolution by influencing how heritable variation arises and persists (Ellegren and Galtier 2016; Charlesworth and Jensen 2021). Testing how chromosome‐scale changes affect gene regulation, cellular traits and heritable variation has long been difficult because such changes are hard to build and evaluate (Bartley et al. 2020; James et al. 2025). High‐throughput synthesis of oligonucleotides and gene fragments, assembly of large DNA molecules and chromosome engineering now allow researchers to write yeast chromosome sequences de novo, nucleotide by nucleotide, or to create new designs by reassembling and restructuring existing DNA (Annaluru et al. 2014; Coradini et al. 2023; Hoose et al. 2023). These approaches make chromosome sequence, gene content and architecture programmable in living cells, bringing broad questions about how genomes give rise to cellular traits and evolutionary change within experimental reach. Work in bacteria first showed that a chemically synthesized genome could be assembled and tested in living cells. In a landmark study, a synthetic bacterial genome was assembled in the budding yeast Saccharomyces cerevisiae and then transplanted into a recipient bacterial cell, where it replaced the native genome and directed cellular function and growth (Gibson et al. 2010). A later study built a reduced bacterial genome through iterative rounds of design, synthesis and testing, revealing a gene set sufficient for cellular life (Hutchison et al. 2016). Together, these studies established genome synthesis as a powerful strategy for biological discovery. However, bacteria lack key features of eukaryotic genomes, including multiple linear chromosomes, nucleosome‐based chromatin and a membrane‐bound nucleus (Venter et al. 2022). Extending this strategy to eukaryotic chromosome biology therefore requires a eukaryotic system in which redesigned chromosomes can be constructed, stably maintained and tested as functional chromosomes in living cells. Among eukaryotes, S. cerevisiae emerged as a natural system for extending chromosome synthesis to eukaryotes (Muller et al. 2012; Archer et al. 2026). This role grew out of earlier work establishing yeast as a host for maintaining and manipulating large DNA molecules. Yeast homologous recombination allows overlapping DNA fragments to be joined in vivo, while defined replication origins, centromeres and telomeres allow the resulting large DNA molecules to replicate and segregate stably (Murray and Szostak 1983; Kunes et al. 1985; Gibson, Benders, Andrews‐Pfannkoch, et al. 2008). These properties made yeast an important platform for cloning, assembling and maintaining chromosome‐scale DNA from diverse organisms, including mammalian genomic regions and phage, bacterial and viral genomes (Burke et al. 1987; Larionov et al. 1996; Gibson, Benders, Andrews‐Pfannkoch, et al. 2008; Jaschke et al. 2012; Thao et al. 2020; Mitchell et al. 2021; Liu et al. 2025). What makes yeast especially powerful as a system for studying synthetic chromosomes, however, is that researchers can replace a native yeast chromosome or chromosomal region with a redesigned version and test its function directly in yeast cells (Annaluru et al. 2014; Coradini et al. 2023). Researchers can therefore alter chromosome sequence, content and architecture and measure the resulting effects on viability, gene regulation and cellular phenotype (Dymond et al. 2011; Shen et al. 2017; Shao et al. 2018; Man et al. 2026). Yeast's compact, well‐annotated genome, extensive community resources and amenability to high‐throughput quantitative and single‐cell phenotyping make the consequences of chromosome redesign easier to interpret (Annaluru et al. 2014; Costanzo et al. 2016; Richardson et al. 2017; Xie et al. 2017; Coradini et al. 2023; Engel et al. 2025). Together, these capabilities frame the central theme of this review: the transition of S. cerevisiae from a host for cloning and assembling chromosome‐scale DNA from other organisms to an experimental synthetic for designing, constructing and testing synthetic versions of its own chromosomes. We first review the technical advances that made yeast a platform for cloning, assembling and maintaining large DNA molecules. We then examine how restructured, chimeric and reduced or minimal yeast chromosomes are being used to test how genome content and organization influence cellular function. Finally, we consider how pairing synthetic chromosome construction with high‐throughput phenotyping and computational prediction could enable iterative design‐build‐test‐learn cycles, moving the field toward a more predictive understanding of how chromosome design shapes cellular traits and evolutionary potential—and toward more reliable engineering of eukaryotic genomes. Past: From Large DNA Molecules to Synthetic Yeast Chromosomes The path to synthetic yeast chromosomes began with the development of methods for maintaining introduced DNA as stable, heritable molecules in yeast. Circular vectors carrying a replication origin derived from the naturally occurring yeast 2 μ plasmid allowed cloned DNA to replicate independently of yeast chromosomes and supported inserts of approximately 5–10 kilobases (Beggs 1978; Jayaram et al. 1983). Researchers next constructed artificial chromosomes containing the elements needed for autonomous replication and stable inheritance, including autonomously replicating sequences and centromeres, allowing much larger DNA molecules to be maintained in yeast (Murray and Szostak 1983; Burke et al. 1987). These yeast artificial chromosomes were subsequently used as vectors to clone and stably maintain foreign DNA fragments of up to several hundred kilobases, including segments of human genomic DNA (Burke et al. 1987). The next advance was to harness yeast homologous recombination to assemble DNA molecules in vivo. Early transformation studies showed that linearized plasmid DNA readily recombined with homologous chromosomal sequences and that double‐strand gaps in introduced plasmids could be repaired using the chromosome as a template (Orr‐Weaver et al. 1981; Orr‐Weaver and Szostak 1983). This recombinational gap‐repair activity was subsequently adapted as a cloning method: cotransforming yeast with a linearized vector and a DNA fragment containing sequences homologous to the vector allowed the two molecules to recombine into a complete circular plasmid in vivo, eliminating the need for prior in vitro ligation (Ma et al. 1987). Transformation‐associated recombination (TAR) cloning extended this homology‐directed cloning strategy to large genomic regions from other organisms by using homology between a vector and a target in high‐molecular‐weight donor DNA, allowing the target region to be selectively captured and maintained in yeast (Larionov et al. 1996; Kouprina and Larionov 2008). Thus, yeast became not only a host for propagating large DNA molecules but also a platform for assembling and selectively cloning them in vivo. Researchers then extended yeast's toolkit from plasmid construction and targeted DNA cloning to the assembly of large DNA molecules from many overlapping fragments. Early studies used yeast homologous recombination to assemble multiple co‐transformed DNA fragments into a single plasmid (Shanks et al. 2006). In 2008, yeast‐based assembly reached genome scale when four quarter‐genome intermediates, built hierarchically from synthetic DNA cassettes, were joined and cloned in S. cerevisiae to produce the 582,970‐bp Mycoplasma genitalium genome, showing that yeast recombination could support the construction of a DNA molecule larger than half a megabase (Gibson, Benders, Andrews‐Pfannkoch, et al. 2008). Follow‐up work showed that the same genome could be assembled from 25 overlapping fragments in a single yeast transformation, underscoring the capacity of yeast to build large DNA molecules from many smaller parts (Gibson, Benders, Axelrod, et al. 2008). Subsequent work used yeast to assemble additional bacterial and viral genomes, as well as large mammalian gene regions, from multiple cotransformed DNA fragments (Jaschke et al. 2012; Fredens et al. 2019; Venetz et al. 2019; Thao et al. 2020; Mitchell et al. 2021; Liu et al. 2025). These advances set the stage for applying yeast‐based DNA assembly to a new target: the yeast genome itself (Figure 1). Through the Synthetic Yeast Genome Project (Sc2.0), researchers began constructing redesigned S. cerevisiae chromosomes de novo from synthetic DNA fragments (Dymond et al. 2011). Early Sc2.0 work established hierarchical strategies for building synthetic chromosome segments from synthetic DNA fragments; the construction of synIII, a synthetic version of Chromosome III, demonstrated that a complete redesigned eukaryotic chromosome could be produced from synthetic DNA (Annaluru et al. 2014). Subsequent Sc2.0 studies extended this construction framework to additional yeast chromosomes, refined methods for their stepwise assembly and sequence correction and developed strategies for consolidating multiple synthetic chromosomes within a single strain (Mitchell et al. 2017; Richardson et al. 2017; Shen et al. 2017; Wu et al. 2017; Xie et al. 2017; Zhang et al. 2017; Zhao et al. 2023; Goold et al. 2025). Synthetic chromosome assembly in yeast. A synthetic chromosome is first designed in silico and divided into overlapping fragments, each of which is produced by de novo synthesis or obtained by molecular cloning. The fragments can then be assembled in vivo by one of two strategies: sequential replacement of the corresponding native chromosome segments using alternating selectable markers, or cotransformation of all fragments with a chromosome backbone carrying a centromere and selectable marker. More recently, complementary approaches expanded the construction toolkit beyond de novo synthesis. CReATiNG uses cloned segments of natural chromosomes, rather than oligonucleotides or synthetic DNA fragments, as programmable building blocks (Coradini et al. 2023). Unique synthetic DNA sequences added to the ends of each segment specify its recombination partners, allowing the cloned segments to be cotransformed and assembled in vivo into a complete synthetic yeast chromosome in a single transformation. Following assembly, researchers can immediately destabilize the native chromosome and select for its loss, thereby rapidly determining whether the synthetic chromosome can support viability (Coradini et al. 2023). The above studies show that complete synthetic yeast chromosomes can now be built through multiple approaches, with control over their sequence, content and organization. Present: Synthetic Yeast Chromosomes as Experimental Systems Building synthetic chromosomes and genomes allows researchers to introduce coordinated perturbations across many genes and chromosomal features and to then test these chromosome‐scale designs in living cells (Coradini et al. 2020; James et al. 2025). Yeast is particularly well suited to such work because it tolerates a wide range of genome manipulations that may be difficult to implement or tolerate in other eukaryotes. For example, yeast cells remain viable when the nuclear genome is consolidated into a single linear or circular chromosome or partitioned among more chromosomes than occur naturally (Sasano et al. 2016; Shao et al. 2018, 2019). As shown in Figure 2, a further critical advantage of yeast is that synthetic chromosomes can be assembled and functionally tested in the same organism (Annaluru et al. 2014; Richardson et al. 2017; Coradini et al. 2023). Together, yeast's tolerance for extensive chromosome‐scale reorganization and its ability to assemble and functionally test synthetic versions of its own chromosomes have opened several dimensions of chromosome design to experimental investigation: Testing synthetic chromosomes in yeast. Synthetic chromosome designs can be assembled while the corresponding native chromosome remains present, allowing construction to be completed before viability is tested. Once assembly of the synthetic chromosome has been confirmed, the native chromosome is destabilized to promote its loss, forcing cells to rely on the synthetic design. Only cells carrying designs that support viability can tolerate loss of the native chromosome. Such strains can then be evaluated for their phenotypes using assays such as growth curves, colony‐size measurements and profiling of transcript, protein or metabolite abundances. Debugging Although chromosome synthesis provides nucleotide‐level control over chromosome sequence and content, both construction errors and intended design features can disrupt function, either individually or through genetic interactions. Sequencing can be used to find unintended substitutions, insertions, deletions and duplications that arise through the DNA synthesis and assembly process, enabling their correction (Xie et al. 2017). For design‐related causes of fitness defects, genetic mapping can be employed to identify the responsible design features, which may arise from the recoding of protein‐coding genes, loxPsym placement, altered tRNA expression, combinations of non‐natural changes on the same synthetic chromosome or different synthetic chromosomes or other factors; modifying these causative elements can restore fitness (Mitchell et al. 2017; Wu et al. 2017; Zhao et al. 2023). These design‐associated failures reveal constraints involving local sequence context, gene expression, genetic interactions among chromosomal elements and potentially other genomic features that are critical for cellular function. Debugging therefore does more than improve synthetic chromosomes; it enables discovery of the sequence and regulatory requirements for chromosome function. Restructuring At the level of individual chromosomes, synthetic approaches have tested how topology and sequence arrangement affect function. Although native S. cerevisiae chromosomes are linear and bounded by telomeres, yeast can tolerate circular derivatives lacking telomeres, altered chromosome endpoints that reposition genes relative to telomeres and reordered genes and chromosome segments (Mitchell and Boeke 2014; Xie et al. 2017; Luo et al. 2018; Coradini et al. 2023). Chromosome architecture can also be varied stochastically: SCRaMbLE uses Cre recombinase and engineered loxPsym sites to generate deletions, duplications, inversions and translocations in Sc2.0 chromosomes, while inducible restriction enzymes and CRISPR‐Cas9 provide additional routes to chromosome restructuring (Dymond et al. 2011; Shen et al. 2016; Muramoto et al. 2018; Fleiss et al. 2019). The resulting variants from these rational and stochastic approaches can be phenotyped and sequenced to link them to cellular traits, revealing that yeast chromosomes can extensively depart from their native architecture (Mercy et al. 2017). Refactoring Because synthetic chromosomes can be designed at the nucleotide level, researchers can remove or modify chromosomal features that complicate engineering and experimentation. Refactoring can include removing repetitive elements and introns, separating overlapping coding regions and modifying local regulatory architecture so that adjacent genes no longer share promoters or terminators (Jaschke et al. 2012; Richardson et al. 2017; Coradini et al. 2020). Such changes can make chromosomes more stable and easier to work with and can also reveal the functions of features that are removed or modified. For example, removal of all 300 spliceosomal introns produced a yeast strain that remained viable but grew slowly because cumulative intron loss caused ribosomal dysregulation, revealing collective effects that were not apparent from individual intron deletions (Man et al. 2026). Minimization Studies of genetically minimal or reduced chromosomes ask how much DNA and how many functional elements can be removed while preserving viability or a defined cellular function. This question has roots in bacterial studies of essential gene sets and genetically minimal cells (Mushegian and Koonin 1996; Glass et al. 2006; Hutchison et al. 2016). In yeast, chromosome‐scale engineering brings this question into a eukaryotic context, where dispensability can depend on genetic background, chromosomal context, growth conditions and interactions among multiple deletions. Rational approaches omit genes or regions predicted to be dispensable and test the resulting chromosomes for viability and function (Coradini et al. 2020; Postma, Couwenberg, et al. 2022). Stochastic approaches, such as Sc2.0's SCRaMbLE, instead generate libraries of compacted chromosomes by deleting nonessential sequences and selecting viable derivatives (Wang et al. 2020; Luo et al. 2021; Malyshava et al. 2026). Viable designs identify sequences that are dispensable under the conditions tested, whereas impaired or inviable designs can reveal hidden dependencies, synthetic‐lethal interactions, dosage requirements and architectural constraints that are not apparent from single‐gene perturbations. Chromosome reduction therefore provides a direct way to define the collective sequence requirements and organisational constraints that support eukaryotic cell function. Chimerism Chimeric chromosomes combine DNA from different strains or species within a single chromosome, allowing the effects of sequence origin to be tested in a common cellular background. By bypassing mating and meiosis, synthetic chromosome assembly can generate defined sequence combinations that may be difficult or impossible to obtain through conventional genetic approaches. Using CReATiNG, Chromosome I segments from two S. cerevisiae strains and S. paradoxus were assembled in all possible syntenic combinations and the resulting strains were phenotyped to determine the individual and interacting effects of the donor segments (Coradini et al. 2023). Synthetic collinearization extends this strategy across greater evolutionary distances: a segment derived from Kluyveromyces marxianus was redesigned to match the gene content and organisation of the corresponding S. cerevisiae region, enabling its installation in S. cerevisiae and revealing incompatibilities between donor proteins and the host proteome (Yang et al. 2025). Modularization The concept of reorganising genes that are normally dispersed across a genome into discrete functional modules was first implemented at genomic scale in the minimal bacterial genome project, in which genes involved in related cellular processes were grouped within a redesigned chromosome segment (Hutchison et al. 2016). In yeast, this strategy has been used to consolidate genes involved in glycolysis and histidine biosynthesis and to relocate tRNA genes to a dedicated neochromosome (Kuijpers et al. 2016; Schindler et al. 2023; Lu et al. 2025). When the corresponding native copies are removed, these modules become the sole source of the encoded function, allowing researchers to determine whether gene sets that are normally distributed across chromosomes remain functional when clustered in a new chromosomal context and to test how alternative alleles or engineered variants affect their collective function. Modular chromosomes can also introduce functions absent from the host, including heterologous metabolic pathways, or consolidate accessory genes drawn from the S. cerevisiae pangenome within a single strain (Kutyna et al. 2022; Postma, Hassing, et al. 2022). Taken together, this work shows how building and testing redesigned chromosomes can turn DNA assembly from a technical achievement into a tool for biological discovery. By varying complementary dimensions of chromosome design, including arrangement, source and retained content, individually or in combination, these approaches generate chromosomes that differ substantially from those found in nature. Testing such designs in cells allows researchers to probe the flexibility and constraints of eukaryotic genome organisation and the functional consequences of sequence combinations that are difficult to generate through conventional genetics. Future: Toward Predictive Design‐Build‐Test‐Learn Cycles The next phase of synthetic chromosome engineering in yeast will depend not only on creating designs that test a broader range of chromosome configurations and biological hypotheses, but also on accelerating their construction and systematically interpreting the outcomes of both successful and unsuccessful designs to improve our ability to predict how chromosome sequence, content and architecture affect viability and cellular phenotype. A major bottleneck is the time required to convert an in silico design into a sequence‐verified chromosome ready for cellular testing, a process that can take months or years depending on chromosome size and assembly strategy (Gibson et al. 2010; Mitchell et al. 2017; Wu et al. 2017; Xie et al. 2017; Fredens et al. 2019; Coradini et al. 2023; Zhao et al. 2023). Errors can arise during DNA synthesis, amplification, propagation, or chromosome assembly, requiring extensive screening and sequencing and, when necessary, further rounds of correction. Not every discrepancy must be corrected, but all must be identified to distinguish the effects of the intended design from those caused by unintended sequence changes. These steps limit both the pace of individual projects and the number of alternative chromosome designs that can be tested. Where these delays arise depends in part on whether assembly is performed in vitro or in yeast. In vitro assembly can be faster and intermediates propagated in E. coli are comparatively easy to sequence‐verify because E. coli grows rapidly and yields abundant plasmid DNA (Gibson, Benders, Andrews‐Pfannkoch, et al. 2008; Gibson et al. 2010; Hutchison et al. 2016). However, in vitro assembly often becomes less efficient as fragment number and construct size increase and reliance on E. coli can impose additional size and stability constraints. Yeast‐based assembly can join many fragments and maintain very large DNA molecules, but recovering and screening transformants can be time‐consuming (Gibson, Benders, Axelrod, et al. 2008; Mitchell et al. 2021; Liu et al. 2025). It can also be impeded by homologous recombination between synthetic fragments and the native yeast genome and sequence verification is particularly challenging when the construct contains sequences identical or highly similar to the host genome (Richardson et al. 2017; Coradini et al. 2023). The optimal approach will depend on the specific project and will continue to evolve as DNA synthesis produces longer, more accurate fragments and both in vitro and yeast‐based assembly methods improve. Regardless of the assembly strategy, construction can be accelerated by reusing cloned natural DNA when de novo synthesis is unnecessary, building chromosome regions in parallel for later consolidation and automating repetitive steps (Mitchell et al. 2017; Coradini et al. 2020; Ma et al. 2024; Goold et al. 2025). Automation could reduce manual effort across the chromosome construction and testing process, including DNA‐fragment handling, in vitro assembly, transformation into E. coli or yeast, colony isolation, screening, sequence verification and phenotyping; recent work has already automated several of these steps (Gartner et al. 2025). Automating these steps could also accelerate debugging and increase the number of designs that can be built and evaluated, allowing researchers to focus more on design, interpretation and learning. Once constructed, designs that appear plausible in silico may prove inviable or produce unexpected phenotypes when tested in cells. At present, many of the constraints underlying these outcomes can be discovered only empirically: although a chromosome can be rationally designed, synthesized or reassembled and introduced into yeast, its viability and function remain uncertain until it is tested in cells. Viability is therefore both a meaningful experimental outcome and the first gate on the path to a functional designer strain. Once viability has been established, the chromosome must be evaluated for the phenotypes expected from its design, with the relevant assays determined by the purpose of the construct. For these outcomes to be interpreted and used for learning, sequence verification and debugging must first establish whether an observed phenotype arises from an unintended construction error or from the intended design. Viable chromosomes should then be evaluated for stable inheritance, fitness and design‐specific molecular and cellular phenotypes across relevant genetic backgrounds and environmental conditions. Yeast is particularly well suited to this work because robotic colony pinning, pooled assays using DNA barcodes and other high‐throughput approaches can evaluate many chromosome designs and conditions in parallel (Giaever et al. 2002; Ehrenreich et al. 2010; Bloom et al. 2013; Giaever and Nislow 2014; Lu et al. 2025). To support interpretation and comparison across experiments, results should be reported for each every tested design, including those that are impaired or nonviable. For each design, researchers should provide the sequence of the assembled chromosome, describe any deviations from the intended design and report important details such as the construction and debugging history. They should also specify the genotype and ploidy of the host strain and the conditions under which that strain was tested. Different classes of synthetic chromosomes, such as those discussed in the preceding section, may reveal distinct and difficult‐to‐predict biological constraints and phenotypic effects, including altered gene regulation or replication dynamics, disrupted chromosomal interactions and unexpected incompatibilities or dependencies among genes and sequence elements (Costanzo et al. 2016; Richardson et al. 2017; Blount et al. 2018; Kuzmin et al. 2018; Luo et al. 2018; Brooks et al. 2022; Kutyna et al. 2022). Once technical causes have been excluded, impaired or nonviable designs are not merely experimental setbacks; they provide evidence of constraints on viable chromosome sequence, content and architecture. Further advances in predictive modelling could accelerate yeast synthetic chromosome studies by screening proposed designs for potential threats to viability, such as dysregulation of key genes, altered biomolecular interactions and changes in chromatin‐associated phenotypes, before they are built (Abramson et al. 2024; Chao et al. 2025; Avsec et al. 2026). Predicting broader phenotypic outcomes is a more distant goal and will likely require multiscale models that integrate replication, segregation, gene dosage, genetic interactions, host background and cellular fitness. Such models could complement faster methods for constructing synthetic yeast chromosomes by focusing experimental effort on both chromosomes predicted to achieve a defined objective and chromosomes whose uncertain viability or phenotype would be especially informative, thereby accelerating successive design‐build‐test‐learn cycles. The usefulness of these models, however, will depend on training data that span the kinds of designs and experimental contexts they are expected to evaluate. Current models are trained largely on naturally occurring genomes, alleles and regulatory states, which represent only a small fraction of the possible sequence and chromosome‐design space (Chao et al. 2025; Schneider et al. 2025; Mandl and Orenstein 2026). They may therefore struggle to predict the behaviour of chromosomes with non‐natural sequences and structures or other features that depart substantially from the training data (Schneider et al. 2025; Mandl and Orenstein 2026). Building synthetic yeast chromosomes offers a way to explore this broader design space experimentally and generate the data needed to improve prediction. However, the construction and phenotyping bottlenecks described above limit the rate at which sufficiently large and diverse training datasets can be generated, thereby slowing model development and successive design‐build‐test‐learn cycles. Despite these limitations, Saccharomyces cerevisiae is particularly well suited to iterative learning because chromosome‐scale designs can be assembled, maintained and functionally tested within the same organism, enabling a design‐build‐test‐learn workflow for eukaryotic genome engineering. In such a workflow, computational models would first evaluate proposed chromosome designs in silico. Selected designs would then be synthesized, reassembled, or generated through direct chromosome engineering and tested in yeast. The resulting data on viability, chromosome stability, fitness, molecular state and cellular phenotype, interpreted in the context of the observed chromosome sequence, host background and growth conditions, would be used to update the models and guide subsequent experiments. Ultimately, prediction should operate in both directions: forecasting the consequences of a proposed chromosome and proposing chromosome designs likely to produce a defined phenotype. Some principles uncovered in yeast may apply broadly across eukaryotes, whereas others will need to be tested and refined in organisms with features that are absent or simplified in yeast, including extensive intron content, alternative splicing, complex regulatory landscapes, developmental programs and higher‐order chromatin organisation. Insights from yeast can therefore provide a foundation for generating and prioritizing hypotheses to be tested in more complex systems. In this way, design‐build‐test‐learn cycles could transform synthetic chromosomes from individual engineering achievements into a scalable framework for uncovering how genome sequence, content and architecture shape biological function and for progressively extending predictive genome design across eukaryotes. Conclusion Synthetic chromosome engineering in yeast is a platform for testing how chromosome sequence, content and architecture shape cellular function. By coupling the design and construction of synthetic yeast chromosomes to systematic cellular testing and modelling, S. cerevisiae can support iterative design‐build‐test‐learn cycles that reveal both alternative chromosome configurations capable of supporting cellular function and the constraints that limit them. In the long run, this knowledge could help move genome engineering toward predictively creating purpose‐designed organisms whose genomes are tailored to defined biological or biotechnological functions that benefit society. Although such designer organisms remain a more distant prospect and principles discovered in yeast will require testing and refinement in more complex systems, research on yeast synthetic chromosomes is helping to establish the experimental and predictive foundation needed to make the design of such organisms increasingly systematic. Author Contributions Daniel T. Lusk: conceptualization, writing – original draft, writing – review and editing. Ian M. Ehrenreich: conceptualization, writing – review and editing, writing – original draft, funding acquisition. Funding This work was supported by the National Institute of General Medical Sciences (R35GM130381), National Science Foundation (2124400), W. M. Keck Foundation, and the University of Southern California. Conflicts of Interest The authors declare no conflicts of interest. Acknowledgements We are grateful to Cara Hull, Zachary Krieger, Elizabeth Moore, Alexandra Christensen, Haily Ne Ville and Jinye Liang for feedback on drafts of this manuscript. Contributor Information Daniel T. Lusk, Email: dlusk@usc.edu. Ian M. Ehrenreich, Email: ian.ehrenreich@usc.edu. Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analysed during the current study. References Associated Data Data Availability Statement Data sharing not applicable to this article as no datasets were generated or analysed during the current study.