Simon Bornschein

"It's pipeline vs platform, Lodge. Pipeline always wins." timmermanreport.com/2026/02/the-pr…

@2048vc IS GROWING! Come join @2048vc team in Boston as an Associate! We are the highest conviction, early stage investor, writing lead checks for pre-seed and seed rounds across Vertical AI, Deep Tech, Health, and Bio. We love being the first YES & we're incredibly privileged to work with founders at the earliest moments in their journeys. Boston has become one of the most exciting places to do that: world-class schools, labs, and a growing community of ambitious founders. So we’re doubling down. This is a rare hire for us, and we are confident this is an exceptional opportunity for someone looking to learn and broaden their super powers in VC. You’ll be on the ground with @thisiszann sourcing, meeting founders, supporting on deals, and helping to build a preeminent pre/seed fund. The sweet spot: 2–5 years of experience, prior venture/early-stage investing exposure, and a strong network. You can be in already in Boston or want to relocate. More information and application in the link below. Please re-share this post with your network and help us spread the word! 👉 2048.vc/bostoninvestor

Pre-Seed should not take months and months to raise! We are thrilled to announce that @2048vc is launching Pre-Seed Fast Track to help you get funding in 10 business days. If you are raising $500K - $1.5M Pre-Seed round in Vertical AI, Deep Tech, Health or Bio we would love to meet you! We offer clear and transparent process and money in the bank within 10 business days of the first meeting. To start the process submit your pitch through our short form - 2048.vc/fasttrack_pitch If there is a potential fit we will reach out within a few days and follow these steps to fund you: Step 1: Intro meeting via Zoom: 30 mins where you meet with 1 member of our team. Step 2: Deep Dive via Zoom: 60 mins where you meet with 3-4 members of our team. Step 3: References + In-Person Meeting: We will chat with your early customers and professional references, and will spend a few hours to get to know each other in person. Startups are long journeys and we know how important the investor-founder fit is, so we will gladly pay for your travel. Step 4: Term sheet: We make you an offer from $250K - $750K Step 5: Wire: We are high conviction and don’t wait for the rest of the round to come together - we wire as soon as we sign the docs. To learn more about Pre-Seed Fast Track visit 2048.vc/fasttrack. We can’t wait to review your pitches and to meet you! @2048vc // #preseed

@BLLPHD @capstantx Congratulations @BLLPHD and team. Really great work and so important for the field.

Ever more true today, so grateful for and proud of the team at @capstantx

Harnessing Preference Optimisation in Protein LMs for Hit Maturation in Cell Therapy • This study explores the use of Protein Language Models (PLMs), specifically ProGen2, fine-tuned via preference optimization for Chimeric Antigen Receptors (CARs) design in cell therapy. The approach aims to enhance the maturation of therapeutic hits by leveraging high-throughput experimental data. • A core innovation: Direct Preference Optimization (DPO) enables fine-tuning by assigning preferences between CAR designs, avoiding the need for extensive labeled datasets. This method aligns PLMs with biological assay outcomes, correlating model loss with T-cell activation performance. • Results show that fine-tuned models achieve a significant correlation (Pearson r > -0.57) between model loss and CAR performance in biological assays, with mutants outperforming baseline CARs by over 2x in certain cases. • The study emphasizes the advantage of preference-driven PLMs in guided exploration of large design spaces, showcasing strong performance in identifying high-activation CAR mutants through exhaustive and greedy search strategies. • Limitations include computational intensity for exhaustive searches and the lack of integration of spatial structure data. Future efforts will explore graph-based learning techniques to address these gaps. • This work sets a new direction for using PLMs in immunotherapy, offering scalable solutions for optimizing CAR designs and paving the way for broader applications in therapeutic development. @nhammerla @coding_bio @SimonBornschein 📜Paper: arxiv.org/abs/2412.01388 #ProteinLanguageModels #Immunotherapy #MachineLearning #CellTherapy #ChimericAntigenReceptors

Who do I know who is a CISO / VP / Director of Cyber Security? Am working with a great early stage company, super credible founders, who are developing tech to neutralise deepfake attacks, starting with vishing. Would love to connect them.

TCR cell therapies vanquish solid tumors — finally nature.com/articles/s4158…

Excited to share our new paper focusing on Second Primary Malignancies (SPMs) following CAR T-cell therapy out today in @CCR_AACR ! 👇 This first-in-kind meta-analysis spans 5,517 lymphoma and myeloma patients receiving a range of CAR-T products. aacrjournals.org/clincancerres/…

Agreed! “While computation alone may have delivered the first fundamental advance in ML in biology, I believe it will not be alone sufficient for the next. The next breakthrough needed for better ML in biology will not be better ML. Rather, it will be better wet-lab methods.”

London central is getting busier with bio…GSK joins Google in London’s Knowledge Quarter thetimes.com/article/6ea896…

Tethered IL15-IL15Rα augments antitumor activity of CD19 CAR-T cells but displays long-term toxicity in an immunocompetent lymphoma mouse model jitc.bmj.com/content/12/7/e…

⚽️🏴󠁧󠁢󠁥󠁮󠁧󠁿⚽️ 🎉🎉🎉

The more I read about antibodies/nanobodies as therapeutic targeting mechanisms (e.g., ADCs, CAR domains), the more I appreciate ML-guided engineering methods that don't seek to maximize affinity, but instead give precise tuning/control over it.

🚀 Calling all Biotech and Digital Health founders! We’re launching our Spring 2024 Studio. Apply to the @2048vc Biotech 🧬and Digital Health 🩺Studio now: get $250K 🤑 for 8% equity in just 16 days. Apply by March 20th to participate in the Spring 2024 Studio! Apply here: 2048.vc/apply_studio We're here to help you build! 💪🔬 Tag a founder who you think should apply below 👇 #Biotech #DigitalHealth #Techbio #Healthtech #Startup #VentureCapital Learn more: 2048.vc/bio_studio and 2048.vc/health_studio

Finetune a protein language model to correctly predict which sequence in a pair has higher fitness. @luoyunan

The target atlas for antibody-drug conjugates across solid cancers | Cancer Gene Therapy nature.com/articles/s4141…

New study in @NatureImmunol unveils how FOXP1 & KLF2 transcription factors intricately balance stem-like and effector phases in CAR T cells, providing insights into improving cell therapies and understanding T cell lineage choices! #CART nature.com/articles/s4159…

@nathanbenaich @spinoutfyi Amazing 🎉🎉🎉

Today’s review marks a huge step forward for UK spinout founders 🇬🇧 It adopts almost all of the policy reform proposals I made via @spinoutfyi and have been pushing on for 2 years. A few thoughts on why this is a big day for the ecosystem and UK science&tech below:

🔬 New #CellTherapy #CART research: CD3δ/ε/γ CAR T cells outperform ζ CARs, revealing new paths for cancer treatment. Enhanced anti-tumor action & reduced T cell dysfunction. #CancerResearch #Immunotherapy Read more: nature.com/articles/s4159…

1/n: We are excited to share that our paper on Chroma, a general purpose diffusion model for proteins, is out today in @Nature! nature.com/articles/s4158… A couple of my favorite highlights in the 🧵below 👇