The complexity of the model systems sharply increases along the drug discovery process, while only a limited number of experimental models are available that accurately reflect human disease.
Conventional in vitro and in vivo models cannot capture disease behavior in real patients, and tools – like CRISPR – don’t act like actual drugs. Drug discovery is costly, time consuming, model systems have poor translation rates to patients and do not significantly reduce the risk of failure in the clinic. This makes it incredibly hard to translate preclinical hypotheses to the clinic and create targeted drugs that truly help.
Before running any wet lab experiments, Turbine computationally simulates tumor cell behavior in patients to understand the complex mechanisms driving the disease. Simulations can reveal the right modality and combination approach to treat even the most resistant cancers. Observing these in silico experiments our biologists and translational experts gain insight into the molecular context by which mono- and combination therapies can potentially lead to patient benefit.
Guiding the R&D process with simulations can increase the chance of success in the clinic.
https://pubmed.ncbi.nlm.nih.gov/14518029/
https://www.nature.com/articles/s41467-021-25175-5
globaldata.com
We use the same “wiring diagram” for all 1600+ in vitro and in vivo models, with patient modeling currently under development.
Our bioplatform allows the setup of cells with distinct OMICS profiles. Behavior training for the models is a special case of Recurrent Neural Networks, using Tensorflow.
Models prepared for simulations are trained on 500.000+ data points (CRISPR, drug sensitivity viability and post-treatment RNASeq assays).
We finish model setup by creating copies with different biomarkers to pinpoint the ideal responders for the specific perturbations.
We can specify extrinsic molecular alterations, to generate a high number of models representative of patients, but non-existent in vitro.
We can introduce interventions at scale, using a versatile toolkit of dose dependent inhibition, interaction level perturbation or combination screens.
Simulating perturbations of models, biomarkers and (even) combinational therapies we generate a complex molecular readout of dose responses and IC50 values.
Filtering and evaluating simulation results enables us to reveal hidden mechanisms behind patient-specific response and its driving effects.
Unique insights on mechanism and the associated biomarkers enable an optimized process of state-of-the-art experimental validation.
Average pathway activity in sensitive and control cells is calculated with a proprietary, footprint-based method based on RNA-sequencing.
We use the same “wiring diagram” for all 1600+ in vitro and in vivo models, with patient modeling currently under development.
Our bioplatform allows the setup of cells with distinct OMICS profiles. Behavior training for the models is a special case of Recurrent Neural Networks, using Tensorflow.
Models prepared for simulations are trained on 500.000+ data points (CRISPR, drug sensitivity viability and post-treatment RNASeq assays).
We finish model setup by creating copies with different biomarkers to pinpoint the ideal responders for the specific perturbations.
We can specify extrinsic molecular alterations, to generate a high number of models representative of patients, but non-existent in vitro.
We can introduce interventions at scale, using a versatile toolkit of dose dependent inhibition, interaction level perturbation or combination screens.
Simulating perturbations of models, biomarkers and (even) combinational therapies we generate a complex molecular readout of dose responses and IC50 values.
Filtering and evaluating simulation results enables us to reveal hidden mechanisms behind patient-specific response and its driving effects.
Unique insights on mechanism and the associated biomarkers enable an optimized process of state-of-the-art experimental validation.
Average pathway activity in sensitive and control cells is calculated with a proprietary, footprint-based method based on RNA-sequencing.
Pioneering an approach that combines simulation with machine learning, we map and model how thousands of signaling proteins interact characterizing cellular level cancer behavior and response or resistance to treatment.
Our platform enables the simulation of drug-like effects from compounds that may not exist yet, on cells potentially unavailable for lab-based testing, like those of high unmet need cancer patients.
This approach will potentially allow us to predict not only what works in cells, mice and people but more importantly, why and how. Continuous iterations of simulations and proprietary in vitro and in vivo experiments confirm predictions and progress our pipeline while simultaneously improving the underlying Simulated Cell™.
As all programs and partnerships run on the latest version of the in silico cell model, training benefits accumulate, leading to a constantly improving platform. Using results to both generate the initial idea and to guide its iterations, as the models improve, this leads to a more rational process to undestand the underlying disease biology.
Our benchmarks show that simulations prevent 2 out of 3 failed experiments in vitro and every 2nd failure in vivo as well.
Simulating cancer behavior to understand drivers of disease & mechanisms of resistance observed in clinical settings.
Uncovering ideal patient population and combination strategy for therapies already in development or on the market.
Identifying truly novel targets to manage unmet need in patients who don’t benefit from existing therapies.
Target ID and Mechanistic profiling
Novel target – biomarker packages for high unmet need patients.
High throughput biological profiling of hit & lead compounds.
Disease positioning
Drug target profiling, indication selection, patient stratification and biomarkers.
Indication expansion
Selection of biomarkers and combinations to overcome resistance and expand indication space.
Building an engineer’s toolkit to understand biology requires software and data scientists who wish to enable biological translational science instead of replacing it.
Our team combines the amazing molecular biology, network- and data science expertise in Hungary with seasoned drug developers from the UK and the US. We’re here to change the status quo and bring truly impactful, simulation-guided therapies to patients.
“Turbine is one of those rare places where a software engineer can have literally lifesaving impact. We work closely with the researchers and witness every day how they use our platform to get better and better predictions. The tech stack is cutting edge: cloud, AI, big data aren’t just buzzwords but practical tools we need to use. And as an added bonus, we get to learn about biology, medicine, and how scientific research works organically – just by being exposed to it constantly!”
“Working here has been quite a journey since day one. Not only the substantial progress we’ve achieved so far, but also personally: as a biologist, I was lured into the field of data analysis. We also have the most inspiring community, with talented and ‘crazy’ people. I’m really a Turbine-addict.”
“Upon joining I immediately started to like Turbine’s multidisciplinary vibe, which I had already got used to during my MSc studies with peers of different professional backgrounds. The fundamental knowledge from my studies proved to be essential for starting as a Cell Developer and later specializing as a Translational Biologist. For me as a social soul, the community was always and still is very important. Just like in the university, many relationships turned into important friendships with my current colleagues.”
“Turbine is an amazing community from various fields and backgrounds, working on the verge of current human knowledge, breaking barriers to succeed where no one else has even attempted to. We do this with relentless curiosity, healthy scepticism, knowing our strength is in our differences, and this inspires me day by day.”
Szabi leads a team of 60+ data scientists, computational and molecular biologists drawn from Budapest’s deep tech talent pool. A serial entrepreneur, Nagy cofounded Turbine in 2015 after launching a cybersecurity startup which was subsequently acquired. Although lacking biotech industry expertise at the time, Szabi has since steered Turbine through multiple top-20 pharma partnerships and financing rounds. Turbine most recently closed a €20M Series A round co-led by Merck’s Global Health Innovation Fund, a unique achievement in Hungary’s small lifesci community.
A computer scientist turned biochemist, Kristóf’s dream is to turn biology into an engineering discipline, enabling science to create human organs as efficiently as we make cars. Named one of the 35 Innovators under 35 in 2018 by MIT Tech Review Europe, Kristóf invented Turbine’s method of simulating human cells and is leading the R&D teams working on the Simulated Cell to make more and more hidden biology available to our internal and external projects.
As a medical doctor specializing in cancer research at some of the world’s top research institutes, Daniel helps design innovative and durable treatments against cancer by converting our platform’s computational results into clinical interpretations. His main responsibility lies in understanding scientific and medical challenges, to then guide our team of biologists, translational and data scientists in generating novel findings to move better treatments to patients.
