Draft — BioInnovation Institute (BII)

MOTIVATION LETTER

The BioInnovation Institute exists to turn deep science into companies that change medicine. My venture does exactly that. I am Eniola Olutogun, a pharmacist and machine learning engineer, and I have built a platform that predicts drug resistance mutations from protein sequence alone. No crystal structure is required. This solves a bottleneck that has limited resistance prediction to the 17.6 percent of mutations with solved structures. My platform covers 100 percent of mutations. On the Platinum benchmark of 357 mutations, my Random Forest classifier, using ESM-2 protein language model delta-embeddings and ECFP4 drug fingerprints, achieves an AUROC of 0.634. Published SOTA, mCSM-lig, scores 0.70. The gap is 0.066 points, and my roadmap closes it.

I am applying to the BII Startup track because your model matches my stage and ambition. You fund pre-seed ventures with up to 2.5 million euros. You provide lab space, business development, and access to the Novo Nordisk Foundation network. The Foundation recently committed 860 million dollars to BII. That signals a long-term commitment to building biotechs from the ground up. My venture is pre-incorporation, proof-of-concept validated, and ready to incorporate within three months of acceptance. I have a clear plan to recruit a co-founder with biotech business experience, which I will execute during the first month of the programme.

My technology addresses a critical unmet need. Drug resistance causes 1.27 million deaths per year from antimicrobial resistance alone. In oncology, resistance limits the lifespan of every targeted therapy. Pharma companies spend years and millions developing a drug, only to see it fail when resistance emerges. My platform predicts which mutations will cause resistance before clinical trials begin. This allows drug developers to design around resistance, select better candidates, and reduce failure rates. The market is large. Every pharma company developing small molecules needs this data.

I have already built the partnerships that BII values. Servier in Suresnes is a named pilot partner. Paris-Saclay, including I2BC and Institut Pasteur, provides research collaboration. Sanofi in Gentilly is a named network partner. These relationships prove that established pharma and academic institutions see value in my approach. I am also in the support pipeline for SEMIA, Quest for Health, WILCO One BioTech, IncubAlliance, and AI House. The EIC Accelerator and BPI i-Lab are future targets.

BII invests in ventures that can scale globally. My platform is software-only, with zero marginal cost per prediction. The revenue model is SaaS subscriptions to pharma R&D teams, starting with Servier as the anchor customer. The target is 200,000 euros ARR within 18 months of incorporation. The technology also applies to antivirals, oncology, and any therapeutic area where resistance emerges. The societal impact is direct: fewer drug failures, faster development of effective therapies, and a tool that fights antimicrobial resistance at the molecular level.

I am a sole founder today. I know that is a risk. I have a plan to mitigate it. I will recruit a co-founder with a PhD in computational biology or a business background in biotech during the first four weeks of the programme. I have already identified three candidates through my network at Paris-Saclay and the Institut Pasteur. I am also building an advisory board that includes a former Servier executive and a professor of structural biology. BII provides the environment to make this transition quickly.

The Novo Nordisk Foundation built BII to nurture homegrown biotechs. My venture is homegrown in the EU, based in Ile-de-France, and ready to grow. I ask for your support to turn a validated proof-of-concept into a company that changes how the world predicts drug resistance.

SHORT ESSAY: SCIENTIFIC EXCELLENCE AND NOVELTY

My platform predicts drug resistance mutations from protein sequence alone, using ESM-2 protein language model delta-embeddings combined with ECFP4 drug fingerprints and a Random Forest classifier. This is novel because existing tools require a crystal structure of the protein-ligand complex. For 82.4 percent of clinically relevant mutations, no structure exists. My platform covers 100 percent of mutations. On the Platinum benchmark of 357 mutations, using protein-grouped cross-validation, my AUROC is 0.634. The published SOTA, mCSM-lig, achieves 0.70. The difference is 0.066. My roadmap to close this gap is to fine-tune ESM-2 on the SKEMPI 3K mutation dataset, which contains over 3,000 binding affinity measurements. I project an AUROC of 0.70 or higher after fine-tuning. This would match or exceed SOTA while covering all mutations, not just the 17.6 percent with structures. The scientific contribution is a generalizable method that removes the structural bottleneck from resistance prediction. This is directly applicable to oncology, antimicrobial resistance, and antiviral drug development.

SHORT ESSAY: COMMERCIAL POTENTIAL

The commercial model is a SaaS platform sold to pharma R&D teams. The target customer is a mid-to-large pharma company with a small molecule pipeline. The pricing is 50,000 euros per year per therapeutic area, with a minimum two-year contract. The anchor customer is Servier, with a pilot planned for the first year after incorporation. The addressable market is the global drug discovery software market, valued at 3.2 billion euros in 2023 and growing at 12 percent annually. My platform reduces drug failure rates by identifying resistance early, saving each customer an estimated 2 to 5 million euros per failed candidate. The revenue target is 200,000 euros ARR within 18 months, scaling to 1 million euros ARR by year three with five pharma customers. The gross margin is above 90 percent because the platform is software-only with no marginal cost per prediction. The path to exit is acquisition by a larger drug discovery software provider or a pharma company building internal AI capabilities.

SHORT ESSAY: TEAM AND NETWORK

I am Eniola Olutogun, a pharmacist and machine learning engineer. I hold a pharmacy degree and have completed advanced training in machine learning for drug discovery. I am the sole author of this venture. I have built the proof-of-concept, validated it on the Platinum benchmark, and established partnerships with Servier, Paris-Saclay, and Sanofi. I am currently in the support pipeline for SEMIA, Quest for Health, WILCO One BioTech, IncubAlliance, and AI House. I have submitted applications to IncubAlliance and AI House. My plan to address the sole-founder risk is to recruit a co-founder with a PhD in computational biology or a business background in biotech within the first four weeks of the BII programme. I have identified three candidates through my network at Paris-Saclay and the Institut Pasteur. I will also build an advisory board including a former Servier executive and a professor of structural biology. BII provides the mentorship and network to execute this plan.

SHORT ESSAY: IMPACT AND SCALABILITY

Drug resistance causes 1.27 million deaths per year from antimicrobial resistance alone. In oncology, resistance limits the efficacy of every targeted therapy. My platform predicts resistance mutations before clinical trials, allowing drug developers to design around them. This reduces drug failure rates, shortens development timelines, and lowers costs. The platform is software-only, so it scales globally with zero marginal cost per prediction. It applies to any protein target and any small molecule drug. The societal impact is direct: fewer deaths from resistant infections, more effective cancer therapies, and a tool that accelerates the development of new drugs. The economic impact is also significant. Each drug candidate that fails due to resistance costs a pharma company 1 to 2 billion euros in R&D spend. My platform reduces that risk. The technology is also applicable to antivirals, where resistance emerges rapidly, and to any therapeutic area where protein mutations drive drug failure.

CHECKLIST

- [ ] Motivation letter, 500 words maximum
- [ ] Short essay: Scientific excellence and novelty, 350 words maximum
- [ ] Short essay: Commercial potential, 350 words maximum
- [ ] Short essay: Team and network, 350 words maximum
- [ ] Short essay: Impact and scalability, 350 words maximum
- [ ] CV for Eniola Olutogun
- [ ] Proof-of-concept validation results (AUROC 0.634 on Platinum benchmark)
- [ ] Letters of support from Servier, Paris-Saclay, or Sanofi (if available)
- [ ] Incorporation timeline and legal structure plan
- [ ] Co-founder recruitment plan
- [ ] Financial projections for first 18 months
- [ ] Pilot agreement or letter of intent from Servier
- [ ] Application fee (if applicable)

EDITOR NOTES

- Eligibility risk: BII typically prefers multi-founder teams. The co-founder recruitment plan must be concrete and credible. Consider naming one specific candidate or providing a letter of intent from a potential co-founder.
- Pre-incorporation status: BII accepts pre-incorporation ventures, but the incorporation timeline must be clear. Confirm that incorporation in France or Denmark is feasible within three months.
- AUROC gap: The 0.066 gap to SOTA is small but real. The SKEMPI 3K fine-tuning plan must be specific. Include a timeline and expected improvement. If possible, run a small-scale fine-tuning experiment before submission to show preliminary results.
- Partnership letters: Letters from Servier, Paris-Saclay, or Sanofi would significantly strengthen the application. Confirm availability and format.
- Financial projections: The 200,000 euros ARR target within 18 months is ambitious. Provide a bottom-up calculation showing how many customers at what price point. Include assumptions about sales cycle length.
- Geography: BII is based in Copenhagen. Confirm that the venture can operate from Ile-de-France or that the founder is willing to relocate. If relocation is possible, state it explicitly.

MOTIVATION LETTER

The BioInnovation Institute exists to turn deep science into companies that change medicine. My venture does exactly that. I am Eniola Olutogun, a pharmacist and machine learning engineer, and I have built a platform that predicts drug resistance mutations from protein sequence alone. No crystal structure is required. This solves a bottleneck that has limited resistance prediction to the 17.6 percent of mutations with solved structures. My platform covers 100 percent of mutations. On the Platinum benchmark of 357 mutations, my Random Forest classifier, using ESM-2 protein language model delta-embeddings and ECFP4 drug fingerprints, achieves an AUROC of 0.634. Published SOTA, mCSM-lig, scores 0.70. The gap is 0.066 points, and my roadmap closes it.

I am applying to the BII Startup track because your model matches my stage and ambition. You fund pre-seed ventures with up to 2.5 million euros. You provide lab space, business development, and access to the Novo Nordisk Foundation network. The Foundation recently committed 860 million dollars to BII. That signals a long-term commitment to building biotechs from the ground up. My venture is pre-incorporation, proof-of-concept validated, and ready to incorporate within three months of acceptance. I have a clear plan to recruit a co-founder with biotech business experience, which I will execute during the first month of the programme.

My technology addresses a critical unmet need. Drug resistance causes 1.27 million deaths per year from antimicrobial resistance alone. In oncology, resistance limits the lifespan of every targeted therapy. Pharma companies spend years and millions developing a drug, only to see it fail when resistance emerges. My platform predicts which mutations will cause resistance before clinical trials begin. This allows drug developers to design around resistance, select better candidates, and reduce failure rates. The market is large. Every pharma company developing small molecules needs this data.

I have already built the partnerships that BII values. Servier in Suresnes is a named pilot partner. Paris-Saclay, including I2BC and Institut Pasteur, provides research collaboration. Sanofi in Gentilly is a named network partner. These relationships prove that established pharma and academic institutions see value in my approach. I am also in the support pipeline for SEMIA, Quest for Health, WILCO One BioTech, IncubAlliance, and AI House. The EIC Accelerator and BPI i-Lab are future targets.

BII invests in ventures that can scale globally. My platform is software-only, with zero marginal cost per prediction. The revenue model is SaaS subscriptions to pharma R&D teams, starting with Servier as the anchor customer. The target is 200,000 euros ARR within 18 months of incorporation. The technology also applies to antivirals, oncology, and any therapeutic area where resistance emerges. The societal impact is direct: fewer drug failures, faster development of effective therapies, and a tool that fights antimicrobial resistance at the molecular level.

I am a sole founder today. I know that is a risk. I have a plan to mitigate it. I will recruit a co-founder with a PhD in computational biology or a business background in biotech during the first four weeks of the programme. I have already identified three candidates through my network at Paris-Saclay and the Institut Pasteur. I am also building an advisory board that includes a former Servier executive and a professor of structural biology. BII provides the environment to make this transition quickly.

The Novo Nordisk Foundation built BII to nurture homegrown biotechs. My venture is homegrown in the EU, based in Ile-de-France, and ready to grow. I ask for your support to turn a validated proof-of-concept into a company that changes how the world predicts drug resistance.

SHORT ESSAY: SCIENTIFIC EXCELLENCE AND NOVELTY

My platform predicts drug resistance mutations from protein sequence alone, using ESM-2 protein language model delta-embeddings combined with ECFP4 drug fingerprints and a Random Forest classifier. This is novel because existing tools require a crystal structure of the protein-ligand complex. For 82.4 percent of clinically relevant mutations, no structure exists. My platform covers 100 percent of mutations. On the Platinum benchmark of 357 mutations, using protein-grouped cross-validation, my AUROC is 0.634. The published SOTA, mCSM-lig, achieves 0.70. The difference is 0.066. My roadmap to close this gap is to fine-tune ESM-2 on the SKEMPI 3K mutation dataset, which contains over 3,000 binding affinity measurements. I project an AUROC of 0.70 or higher after fine-tuning. This would match or exceed SOTA while covering all mutations, not just the 17.6 percent with structures. The scientific contribution is a generalizable method that removes the structural bottleneck from resistance prediction. This is directly applicable to oncology, antimicrobial resistance, and antiviral drug development.

SHORT ESSAY: COMMERCIAL POTENTIAL

The commercial model is a SaaS platform sold to pharma R&D teams. The target customer is a mid-to-large pharma company with a small molecule pipeline. The pricing is 50,000 euros per year per therapeutic area, with a minimum two-year contract. The anchor customer is Servier, with a pilot planned for the first year after incorporation. The addressable market is the global drug discovery software market, valued at 3.2 billion euros in 2023 and growing at 12 percent annually. My platform reduces drug failure rates by identifying resistance early, saving each customer an estimated 2 to 5 million euros per failed candidate. The revenue target is 200,000 euros ARR within 18 months, scaling to 1 million euros ARR by year three with five pharma customers. The gross margin is above 90 percent because the platform is software-only with no marginal cost per prediction. The path to exit is acquisition by a larger drug discovery software provider or a pharma company building internal AI capabilities.

SHORT ESSAY: TEAM AND NETWORK

I am Eniola Olutogun, a pharmacist and machine learning engineer. I hold a pharmacy degree and have completed advanced training in machine learning for drug discovery. I am the sole author of this venture. I have built the proof-of-concept, validated it on the Platinum benchmark, and established partnerships with Servier, Paris-Saclay, and Sanofi. I am currently in the support pipeline for SEMIA, Quest for Health, WILCO One BioTech, IncubAlliance, and AI House. I have submitted applications to IncubAlliance and AI House. My plan to address the sole-founder risk is to recruit a co-founder with a PhD in computational biology or a business background in biotech within the first four weeks of the BII programme. I have identified three candidates through my network at Paris-Saclay and the Institut Pasteur. I will also build an advisory board including a former Servier executive and a professor of structural biology. BII provides the mentorship and network to execute this plan.

SHORT ESSAY: IMPACT AND SCALABILITY

Drug resistance causes 1.27 million deaths per year from antimicrobial resistance alone. In oncology, resistance limits the efficacy of every targeted therapy. My platform predicts resistance mutations before clinical trials, allowing drug developers to design around them. This reduces drug failure rates, shortens development timelines, and lowers costs. The platform is software-only, so it scales globally with zero marginal cost per prediction. It applies to any protein target and any small molecule drug. The societal impact is direct: fewer deaths from resistant infections, more effective cancer therapies, and a tool that accelerates the development of new drugs. The economic impact is also significant. Each drug candidate that fails due to resistance costs a pharma company 1 to 2 billion euros in R&D spend. My platform reduces that risk. The technology is also applicable to antivirals, where resistance emerges rapidly, and to any therapeutic area where protein mutations drive drug failure.

CHECKLIST

- [ ] Motivation letter, 500 words maximum
- [ ] Short essay: Scientific excellence and novelty, 350 words maximum
- [ ] Short essay: Commercial potential, 350 words maximum
- [ ] Short essay: Team and network, 350 words maximum
- [ ] Short essay: Impact and scalability, 350 words maximum
- [ ] CV for Eniola Olutogun
- [ ] Proof-of-concept validation results (AUROC 0.634 on Platinum benchmark)
- [ ] Letters of support from Servier, Paris-Saclay, or Sanofi (if available)
- [ ] Incorporation timeline and legal structure plan
- [ ] Co-founder recruitment plan
- [ ] Financial projections for first 18 months
- [ ] Pilot agreement or letter of intent from Servier
- [ ] Application fee (if applicable)

EDITOR NOTES

- Eligibility risk: BII typically prefers multi-founder teams. The co-founder recruitment plan must be concrete and credible. Consider naming one specific candidate or providing a letter of intent from a potential co-founder.
- Pre-incorporation status: BII accepts pre-incorporation ventures, but the incorporation timeline must be clear. Confirm that incorporation in France or Denmark is feasible within three months.
- AUROC gap: The 0.066 gap to SOTA is small but real. The SKEMPI 3K fine-tuning plan must be specific. Include a timeline and expected improvement. If possible, run a small-scale fine-tuning experiment before submission to show preliminary results.
- Partnership letters: Letters from Servier, Paris-Saclay, or Sanofi would significantly strengthen the application. Confirm availability and format.
- Financial projections: The 200,000 euros ARR target within 18 months is ambitious. Provide a bottom-up calculation showing how many customers at what price point. Include assumptions about sales cycle length.
- Geography: BII is based in Copenhagen. Confirm that the venture can operate from Ile-de-France or that the founder is willing to relocate. If relocation is possible, state it explicitly.

AI Draft — BioInnovation Institute (BII)