Machine Learning for Strain Selection
Applying predictive models to phenotypic and genomic data to prioritize fungal strains for yield, quality or bioprocess performance.
Program identification
- 1.1Program name
- Machine Learning for Strain Selection
- 1.2Type of qualification
- Course
- 1.3Provider / endorsing institution
- Fungaris – International Institute of Mycology, School of Digital Mycology: AI, Genomics & Data. A Fungaris program of its own, not endorsed by a university.
- 1.4Mode of study
- Online, with live sessions in a small group, recordings and a virtual campus.
- 1.5Duration (hours / ECTS credits)
- 3 months (12 weeks) · 72 hours in total · 6 modules of 12 hours · 12 live sessions of 2 hours. ECTS credits: not applicable (Fungaris own program, not an official university degree).
- 1.6Language
- English
- 1.7Start and end dates / intake
- Next intake: January 11, 2027 to April 2, 2027. Enrollment closes on January 4, 2027. Following intake: March 8, 2027. How intakes work →
Overview
2.1Program description
Machine Learning for Strain Selection treats choosing the best fungal strains as a prediction problem. You learn how to collect phenotypic data at scale, from growth rates to yield and quality traits, and how to apply supervised regression and classification models. Genomic prediction and marker-assisted selection show how DNA data can complement phenotypes, while dedicated modules tackle small datasets, overfitting and validation, and active learning to decide which strains to test next.
Strain banks and breeding programs often hold far more candidates than they can test, so smarter prioritization saves time, space and money. Machine learning can help, but only when data is well structured and models are honestly validated. The course suits breeders, strain bank curators and biotech scientists with basic data analysis skills. For the final project you build and validate a model on supplied or your own strain data and propose a prioritized screening list with clearly stated uncertainty.
2.2Why it matters / what need it meets
It meets a practical need: applied, up-to-date knowledge in digital mycology, useful in areas such as mushroom breeding and spawn production, industrial biotechnology, culture collections, food and mycoprotein production.
Mycology is moving fast and good practical training is still scarce. This program gives you a structured path, taught live by people who work in the field, so you can apply what you learn with judgment and safety.
Objectives and competences
3.1General objective
To finish the program able to apply its content professionally and with judgment. In short:
Applying predictive models to phenotypic and genomic data to prioritize fungal strains for yield, quality or bioprocess performance.
3.2Specific objectives
- Build the foundations: Strain selection as a prediction problem; Phenotyping and data collection at scale.
- Put them into practice: Supervised learning: regression and classification; Genomic prediction and marker-assisted selection.
- Apply everything to a real case: Small data, overfitting and validation; Active learning for screening campaigns.
- Complete a final project reviewed by your instructor.
3.3Competences you will gain
- Frame strain selection goals as supervised prediction tasks.
- Organize phenotypic and genotypic strain data for modeling.
- Train regression and classification models with scikit-learn or R.
- Validate models rigorously on small datasets to avoid overfitting.
- Use active learning to prioritize strains for the next screening round.
Student profile
4.1Who it is for (entry profile)
Breeders, strain banks and biotech scientists.
4.2Entry requirements
Basic programming in Python or R and introductory statistics are expected; prior machine learning experience is not required.
4.3Career paths (exit profile)
- Data scientist in a fungal breeding program
- Strain bank curator using predictive tools
- Biotech scientist in strain development
- R&D analyst for a spawn producer
- Computational biologist in industrial mycology
Outcomes depend on your experience and on your local market.
Study plan
5.1Overall structure (modules and workload)
| Module | Title | Phase | Topics | Hours |
|---|---|---|---|---|
| 01 | Strain selection as a prediction problem | Foundations | 3 | 12 |
| 02 | Phenotyping and data collection at scale | Foundations | 3 | 12 |
| 03 | Supervised learning: regression and classification | Practice | 3 | 12 |
| 04 | Genomic prediction and marker-assisted selection | Practice | 3 | 12 |
| 05 | Small data, overfitting and validation | Application & project | 3 | 12 |
| 06 | Active learning for screening campaigns | Application & project | 3 | 12 |
| Total | 72 |
6 modules × 12 hours = 72 hours: 24 hours of live sessions, and the rest for campus study, exercises and the final project.
5.2Content by module
Hover or tap a module to follow the path from the first session to the final project. Open a module to see the topics it covers.
- Breeding and selection goals
- Framing selection as prediction
- Choosing target traits
- High-throughput growth assays
- Image-based phenotyping
- Structuring strain databases
- Linear and regularized regression
- Random forests and gradient boosting
- Classification of strain categories
- Genotyping strains
- Genomic prediction models
- Marker-assisted selection in fungi
- Cross-validation strategies
- Overfitting with few strains
- Reporting model uncertainty
- Exploration versus exploitation
- Uncertainty-based selection
- Planning iterative screening rounds
5.3Company placement
This program does not include a company placement. Practice is built into every module through exercises, case studies and the final project.
5.4Final project
A validated predictive model for strain performance with a prioritized screening list and uncertainty estimates.
Methodology
6.1How it is taught
- Live online sessions, usually weekly, with a working professional and a small group.
- Recordings of every session, so you can catch up if work or time zones get in the way.
- Exercises and real case studies in every module.
- Tutoring: your instructor answers questions and gives feedback on your exercises and project.
6.2Platform and resources
A Moodle virtual campus with session recordings, readings, exercises and forums for each module. You need a computer, an up-to-date browser and a stable connection; a webcam and microphone are recommended.
Technical requirements →6.3Estimated weekly workload
About 6 hours a week for 12 weeks: one 2-hour live session on a weekday evening, plus study and exercises on the campus at your own pace.
Assessment
7.1Assessment system and weighting
Progress is assessed through participation in live sessions, modeling exercises in each module and a final project reviewed by the instructor.
7.2Requirements to pass
- Complete every module, by attending the live sessions or watching the recordings.
- Hand in the module exercises.
- Have your final project approved by your instructor.
Teaching team
8.1Academic direction
At Fungaris, the Digital Mycology school coordinates the content and quality of this program. The academic director of each intake is named in the intake guide.
8.2Faculty
Taught by professionals who work in the field. The instructor for each intake is confirmed before the start.
Meet our faculty →Certificate
9.1Certificate awarded
Course Certificate
On completing the program and its project, Fungaris issues a Course Certificate. Fungaris certificates are the institute's own. Whether an employer, school or authority recognizes them depends on that organization and on local rules in each country.
9.2Issuing body / recognition
Issued by Fungaris (operated by Nexora Generation, S.L.) as a digital certificate. It is a Fungaris certificate of its own: it is not an official university degree and does not award ECTS credits.
About our certificates →Admission and enrollment
10.1Admission process (steps)
- Send your request through the form or write to admissions.
- Admissions reviews your profile and confirms the start date, the session schedule and the fee.
- You secure your place with full payment or a 40% reservation.
- You receive your campus access and welcome guide before the first session.
10.2Documents required
- A valid identity document.
- Contact and billing details (personal or company).
10.3Deadlines
Enrollment for the intake starting on January 11, 2027 is open until January 4, 2027, or until the group is full (maximum 20 students).
10.4Fee and payment options
Fee: $450
- Full payment before the start: $360 (20% reduction).
- 40% reservation ($180) and 3 monthly installments of $90.
10.5Grants, discounts and funding
Discounts: 10% for Fungaris alumni and 15% for groups of three or more people from the same organization. Discounts cannot be combined with each other or with the full-payment reduction.
Your employer can pay for the program and receive the invoice in the company name. Whether a public or employer training budget can cover the fee depends on your country, so check with them.
10.6Cancellation policy
The conditions for canceling, withdrawing or moving to a later intake are explained in our cancellation policy and in your enrollment confirmation. Nothing there limits your rights under the consumer law that applies to you.
Cancellation and refunds →Additional information
11.1Frequently asked questions
Do I need experience to study Machine Learning for Strain Selection?
Basic programming in Python or R and introductory statistics are expected; prior machine learning experience is not required.
How is Machine Learning for Strain Selection taught?
Live online, in 6 modules with a working professional, a small group and a final project.
How am I assessed?
Progress is assessed through participation in live sessions, modeling exercises in each module and a final project reviewed by the instructor.
What is the final project?
A validated predictive model for strain performance with a prioritized screening list and uncertainty estimates.
What certificate do I get?
On completing the program and its project, Fungaris issues a Course Certificate. Fungaris certificates are the institute's own. Whether an employer, school or authority recognizes them depends on that organization and on local rules in each country.
What can I study after Machine Learning for Strain Selection?
The natural next step is Diploma in Fungal Bioinformatics.
11.2Alumni testimonials
We only publish verified testimonials from real students. This program does not have any published yet; they will appear here once its first intakes finish.
11.3Admissions contact
Write to info@fungaris.institute or use the contact form. Admissions answers in your language.
Keep branching.
Join the next intake.
Pay in full with a 20% reduction, or reserve your place with 40% and pay the rest in monthly installments. All prices are stated in US dollars (USD).