Edinburgh PhD: Evolution of Structural Variation through Pangenomics

Edinburgh PhD: Evolution of Structural Variation through Pangenomics

Research question and application timing

This Edinburgh project is an entirely computational evolutionary-genomics doctorate: analyse a new Danaus chrysippus butterfly pangenome alongside thousands of published lepidopteran genomes, develop software, and optionally use simulation. Its scientific value extends beyond one butterfly: it asks how structural variants originate, affect recombination and genes, and change association mapping. The advertised mentors are Simon Martin and Konrad Lohse.

The project instructions give a deadline of midday on Monday 16 November 2026 (timezone not stated), 1 October 2027 entry and four-year competitive EASTBIO/Edinburgh Doctoral College funding. Whether applications remain open is unconfirmed.

The university project directory lists the exact title, institution, both supervisors, deadline and worldwide eligibility.

The linked eligibility checklist still needs confirmation.

Martin's general opportunities page says “No positions currently available”; clarify its relationship to this specific listing and actual supervisory capacity rather than treating it as proof of expiry. Official directory entry point, Lab wording, School PhD funding information.

Chinese international applicants can compete for EASTBIO, with a fee waiver for successful international award-holders and a 30% annual international cap. That cap concerns EASTBIO's annual UKRI-funded intake, not the separate EDC route or an individual's probability. EASTBIO eligibility and fee-waiver policy.

Research Problem and Doctoral Work

The Four Questions Must Remain Distinct

Structural variants include insertions/deletions and larger rearrangements. Single-reference detection and a focus on already conspicuous trait/disease effects can bias what researchers discover. Multiple assembled genomes allow a broader starting point, but a pangenome does not automatically remove assembly errors, population sampling bias or uncertainty about functional effects.

The advertised doctoral tasks address four separate questions:

1. What determines the rate at which structural variants evolve?

2. How do different variants alter recombination?

3. What happens to genes inside rearranged regions?

4. How does pangenomics affect GWAS accuracy?

These require different comparisons. More detected variants alone does not establish improved association accuracy, and a statistical association does not demonstrate a gene’s mechanism. The new butterfly data and existing comparative genomes should support a coherent thesis rather than four unrelated catalogues.

Large genomic analysis and software development are explicit work; simulation is optional. No wet-lab component is required by this advert. The lab undertakes broader breeding/field activities, but those should not be silently added to this computational position. Lab research scope.

What a Credible Computational Contribution Might Involve

A defensible project would document data provenance and assembly quality, define comparable structural-variant categories across genomes, connect variants to recombination and gene-content evidence, and evaluate association results against appropriate baselines. These are analytical priorities, not an announced software stack or fixed pipeline. Ask which new assemblies, population samples and phenotypes already exist, and what ground truth or sensitivity analyses are realistic.

The 2026 Fang and Edwards review describes how multiple high-quality genomes reveal sequence/gene-level variation missed by single references and notes that animal pangenomics remains relatively young. It supports the field’s momentum without establishing that every graph representation or sampling design is unbiased. Author-lab review abstract.

Possible outputs include reusable analysis software, comparative results and an evolutionary explanation of structural variation; a particular public release, benchmark size or accuracy gain is not promised. Confirm software licensing, documentation/testing expectations, authorship, data embargoes and a viable first-year deliverable before treating open tools as an assured outcome.

School, Programme and Field Context

The host and degree-awarding institution is the University of Edinburgh, School of Biological Sciences, with work in the Institute of Ecology and Evolution at Ashworth Laboratories. This is a biology/evolutionary-genomics registration rather than a Computer Science degree, despite the software emphasis. The lab’s official contact locates it in Edinburgh, UK. Host and research location.

The school reports fifth in the UK for REF 2021 research power, a quality-and-breadth measure, and 94% world-leading/internationally excellent research. The news heading’s shorthand “research quality” should not replace the stated power metric. Its separate school introduction reports QS Biological Sciences 2025 top 25 worldwide. These are subject/school signals, not a ranking of this computational project or mentoring quality. REF scope, School environment and subject ranking.

The advertised route is PhD Biological Sciences with Internship, four-year full-time for 1 October 2027 entry. The 2027 programme sets a first-year probationary period, a thesis committee including an independent chair, 10-week and 10-month reports and annual reviews. It describes a year-two poster, year-three oral presentation, a thesis of at most 100,000 words and a viva.

The internship is ordinarily three months; approval for up to six months does not itself extend funding.

The part-time heading (eight years) and body (84 months) conflict; the advertised funded route remains four-year full-time.

EASTBIO visa-requiring students must register full-time and live within commuting distance except authorised fieldwork/placements. Current 2027 programme.

Supervisors, Papers and Research Resources

Martin studies genomic diversity, adaptation, speciation and genome organisation. The lab lists existing doctoral work on rearrangements, insertions/deletions and pangenomes; one pangenome student is co-supervised by Lohse. This demonstrates a relevant training environment, while also making the division between an existing student’s project and this new position worth clarifying. Current lab people.

Lohse’s institutional record identifies him as a Senior Lecturer, with current population-genomic software and inference research. His group studies selection, speciation and recombination through computational methods. A separately advertised postdoctoral job is not proof of doctoral capacity. Meeting frequency, daily coding support and the mentors’ division of work remain unconfirmed. Institutional profile, Lohse lab.

Three Recent Papers and Their Limits

• De-Kayne and colleagues, 2025, PLOS Biology: Martin’s coauthored study analyses 174 butterflies, finds at least six divergent haplotype groups in the BC supergene, and detects natural recombination despite suppression. It gives direct biological reasons to distinguish rearrangement modules and residual gene exchange. This is one supergene/population study, not proof that every rearrangement behaves alike or the advertised new pangenome is ready. Primary paper.

• Laetsch and colleagues, 2023, gIMble, PLOS Genetics: including both advertised supervisors, it integrates demographic inference and genomic barrier analysis with an isolation-with-migration framework and software. The relevance is explicit modelling of alternative processes and reusable computational work. Its assumptions still matter; it is not an unbiased structural-variant detector or a substitute for pangenome validation. Primary paper.

• Setter, Lohse and Baird, 2026, diempy, G3: an efficient Python implementation supports reference-panel-free genome polarisation and chromosome painting, with ploidy-aware processing and visualisation. “Polarisation” here assigns variants to opposite sides of a gene-flow barrier, not ancestral/derived states. It evidences active software expertise, not a complete graph-pangenome assembler or a 2027 supervision commitment. Publication record and abstract.

The project also cites Kim 2022 on stepwise supergene evolution and Wood 2026 on an ash pangenome as background only; no additional scientific conclusions are drawn from them. Kim reference, Wood reference.

Complete publication records and reading scope

These are three distinct recent peer-reviewed original studies. Their summaries rely on abstracts and selected substantive results. Publication details and author order/roles were checked against the bibliographic records listed below; these checks do not amount to a full-text review of methods, supplementary material or underlying data. No additional full-paper reading is claimed.

The independent 2026 Fang/Edwards review is field context, not a fourth supervisor original.

• Rishi De-Kayne; Ian J. Gordon; Reinier F. Terblanche; Steve Collins; Kennedy Saitoti Omufwoko; Dino J. Martins; Simon H. Martin. Incomplete recombination suppression fuels extensive haplotype diversity in a butterfly colour pattern supergene. PLOS Biology 23(2):e3003043. DOI: 10.1371/journal.pbio.3003043. Online 2025-02-28; issue February 2025. Martin seventh/last, co-corresponding with De-Kayne link 1

• Dominik R. Laetsch; Gertjan Bisschop; Simon H. Martin; Simon Aeschbacher; Derek Setter; Konrad Lohse. Demographically explicit scans for barriers to gene flow using gIMble. PLOS Genetics 19(10):e1010999. DOI: 10.1371/journal.pgen.1010999. Online 2023-10-10; issue October 2023. Martin third/six; Lohse sixth/last and corresponding; Laetsch/Bisschop equal first contribution link 1 link 2

• Derek Setter; Konrad Lohse; Stuart J. E. Baird. diempy: fast and reference-free genome polarization and chromosome painting. G3: Genes, Genomes, Genetics 16(8):jkag140. DOI: 10.1093/g3journal/jkag140. Online 2026-06-01; issue August 2026. Lohse second/three; exact contribution/correspondence not resolved from university metadata; do not assume last/lead link 1 link 2

The institutional index describes the Eddie high-performance computing service but does not establish this student’s CPU/RAM, storage, priority allocation, backup or software support. These remain material feasibility questions for many-genome analysis. Eddie high-performance computing information.

Funding, Application and Living Cost

Which Award and Which Budget?

The project information offers consideration for either EASTBIO BBSRC Doctoral Landscape Award or Edinburgh Doctoral College Award, with four-year support, tuition cover, UKRI-rate monthly maintenance, project costs and mandatory training including a three-month placement. No separate application is required for the college award.

The School’s funding information says the college award is incorporated into EASTBIO recruitment, with a project grant of £5,000 in years 1–3 and £1,000 in year 4. This is research expenditure, not personal maintenance, and should not be replaced with another host’s fourth-year budget.

Central EDC materials retain older 2026/27 terms, and 2027/28 status is unconfirmed; neither establishes this project’s final award route or amount.

Confirm the actual route and final award conditions rather than silently resolving the school/central mismatch. Central scheme status reference.

EASTBIO rules confirm international eligibility, fee top-up waiver and exclusion of visa/IHS costs. Local-residence and visa/full-time requirements matter. Confirm equivalent terms for a college-funded associate award. The UKRI minimum published for 2026/27 is £21,805 from October 2026, with Home-fee support £5,238 for 2026/27; neither establishes this project’s 2027/28 stipend. Current UKRI reference.

Applying: Preserve the Host-Specific Timing

The project requires Euclid, 2027/28 Biological Sciences with Internship, start 1 October 2027, with title, supervisors, funding source and the completed EASTBIO form uploaded in the Research Proposal section. That upload label does not mean an additional free-form proposal is required.

References use the official template; the advert allows two weeks after the midday 16 November cutoff, while the general EASTBIO guidance says references are due 16 November. Resolve this before relying on the extension.

EASTBIO guidance additionally requires the anonymous EDI response ID, academic transcripts and two academic/professional references; project supervisors cannot act as referees. Only one EASTBIO project may be selected across hosts. Contact the mentor before applying. Extra CVs are disregarded in the formal EASTBIO competition. Current application rules.

The 2027 selection timeline lists supervisor nominations 18 January, invites 22 February, interviews in the week beginning 8 March, and decisions in the week beginning 22 March. The host may request missing items through Euclid; check it and email regularly. Scholarship changes/withdrawal remain possible. Selection criteria and dates, Current recruitment contact.

Edinburgh Housing and Maintenance

The university’s published 2026/27 postgraduate budget is £1,542 monthly / £18,504 over 52 weeks. Monthly components are accommodation/utilities £978, food/shopping £248, phone £12, leisure/wellbeing £246 and bus £58. Accommodation-type examples span £752–£1,249 monthly; rent, utilities and laundry inclusion differ. Edinburgh 2026/27 postgraduate living-cost budget.

The official cost and stipend references do not establish a personal disposable balance for 2027. Budget separately for visa/IHS, relocation, deposits, study costs and future price changes. Use twelve months and the actual housing contract; family and short-lease budgets differ.

Applicant Fit and Career Implications

EASTBIO eligibility requires a relevant 2:1 or international equivalent, with no mandatory master’s; nonstandard backgrounds may be considered. The 2027 programme requires IELTS Academic 6.5 overall and 6.0 each, without One Skill Retake. TOEFL taken before 21 January 2026 requires 92 overall / 23 in each element; from that date 4.5 overall / 4.5 in each element, without MyBest. IELTS/TOEFL are normally valid within two years of the first day of the starting month.

The programme has no application fee.

The programme also lists other accepted tests with route-specific scores and validity.

An English-taught degree is an alternative only where the stated country/institution and recency rules are met; a Chinese English-medium degree is not an automatic waiver.

For non-majority-English-country nationals an eligible English-taught and assessed degree must be no more than five years old on the first day of the entry month.

The current China selector does not publish a universal percentage threshold for this PhD: the 80–85% table applies to taught postgraduate courses and two named MScR programmes.

Obtain an individual qualification assessment. Current academic/English requirements.

The advert does not prescribe a programming language. Useful evidence would show reproducible genomic analysis, readable software, statistical reasoning and understanding of reference/assembly bias. Simulation should remain optional. Applicants seeking primarily wet-lab work or a generic AI application may find the advertised evolutionary-inference/software emphasis different from their aims.

Career directions

Programming, data quality, modelling and communication are transferable beyond butterfly research. Government data-science guidance supports these skill categories; it does not establish project-specific graduate destinations, salary or automatic eligibility for every commercial role. Career evidence.

Next Actions and Questions to Verify

Before Applying

1. Clarify the lab opportunities page’s generic wording and Martin/Lohse’s actual supervision capacity for the project listing.

2. Resolve the midday cutoff timezone and reference-extension conflict; prepare the official form, EDI ID and host portal documents.

3. Verify Chinese qualification equivalence, 2027 English rules, the single-project choice and funding-route eligibility.

At Interview

1. Which new assemblies, samples and phenotypes are ready, and what benchmarks validate structural calls and GWAS comparisons?

2. What is the division of work across all four scientific questions, software and optional simulation?

3. What CPU/RAM/storage allocation, data rights, software licensing, coding supervision and first-year milestones are committed?

4. How is overlap with current students managed, and where will the mandatory placement fit?

Before Accepting

1. Obtain written 2027 stipend, full international tuition, exact award route, project budget, duration and interruption/extension terms.

2. Confirm registration/progression, placement funding, visa/IHS and a twelve-month housing budget.

3. Keep the project listing, admission decision and individual funding award distinct.

Evidence and Sources

The paper-specific evidence limits are stated above. The saved materials do not establish whether the vacancy is still accepting applications, whether the student will have access to individual datasets, or whether any applicant will receive a 2027 award.

Project, school and funding

• School 2027 PhD funding

• EASTBIO 2027 recruitment

• EASTBIO 2027 application guidance

• EASTBIO eligibility and funding FAQ

• EASTBIO 2027 selection timeline

• REF 2021 biology school results

• School research environment

• UKRI minimum-stipend guidance

• Internship PhD programme reference

• Edinburgh 2026/27 postgraduate living-cost budget

• Central EDC scheme status reference

Supervisors, papers and computing

• Martin lab research

• Martin lab people

• Martin lab opportunities and capacity caveat

• Konrad Lohse institutional profile

• Lohse lab

• Incomplete recombination suppression fuels extensive haplotype diversity in a butterfly colour pattern supergene

• Demographically explicit scans for barriers to gene flow using gIMble

• diempy: fast and reference-free genome polarization and chromosome painting

• Fang and Edwards 2026 pangenomics review

• Eddie high-performance computing information

• Kim 2022 supergene background as listed

• Wood 2026 ash background as listed

Career context

• Government data-science career context