About the role
As a chemical data scientist at Treeline, you will play a key role in building, validating, and deploying machine learning models to drive the design of oncology therapeutics. To ensure optimal predictive value you will contribute to data collection and acquisition strategy, tightly collaborating with computational and medicinal chemists, structural biologists, biophysicists, and other data scientists to assess druggability, guide synthesis decisions, and design experiments. Additionally, you will drive advances in technology and infrastructure through internal and external engagement to build a world-class data science environment that accelerates the path to medicines, including improvements to the human-machine interface.
Watertown, MA; Alternate locations will be considered including San Diego, CA, and the Philadelphia/New York metropolitan areas.
- Collaborate with multi-disciplinary teams in the design of medicines.
- Build quantitative structure-activity relationship (QSAR) models for ADMET, physicochemical properties, selectivity, and potency.
- Rigorously validate models with the appropriate metrics.
- Deploy models into a modern service-based cloud infrastructure.
- Integrate active learning into the design-make-test cycle
- Maintain software, submit code patches upstream to open-source software as needed
- Drive development of tools that allow scientists to take optimal advantage of experimental data analysis and computational methodologies.
- Ensure that user interfaces to data and predictive science approaches are ergonomically designed.
- Survey the literature for novel cheminformatics and QSAR methodology and algorithms, including deep neural networks and multitask learning, and introduce to the broader team when appropriate.
- Work collaboratively with the Treeline team to apply and enhance computational infrastructure needed to drive our projects forward.
- Contribute to recruiting of exceptional talent.
- Manage vendor relationships, engage in external collaborations, and set a high standard for internal software engineering.
- PhD or equivalent in Computer Science, Machine Learning, Artificial Intelligence, Cheminformatics, Computational Chemistry, or related quantitative fields with 2+ years post-PhD experience in the Pharmaceutical/Biotechnology Industry (or MS in one of these fields with 7+ years of related experience).
- Strong understanding of chemistry, biological assays, drug metabolism, pharmacokinetics, pharmacodynamics, toxicology, and statistics.
- Extensive experience developing and impactfully applying QSAR models to drug discovery.
- Hands on experience with data aggregation, manipulation, integration, mining, visualization and analysis, including structured and unstructured data sources.
- Desire and ability to dig deeply through various data types to gain insights and develop hypotheses.
- Experience with machine learning techniques such as reinforcement learning, meta learning, active learning, generative models, regression models, binary classifiers, including application to medicinal chemistry problems.
- Experience with small molecule chemistry toolkits: DeepChem, RDKit, OEChem, etc, as well as PyTorch, TensorFlow, Theano.
- Extensive experience working in command line interfaces and fluency in scripting programming languages e.g., scientific Python (numpy, scipy, scikit-learn).
- Proven ability to influence across multi-disciplinary teams and interact with external experts in academic institutions as well as CROs.
- Excellent written and oral communication skills, including an ability to explain complex ideas to computational scientists, experimentalists, and clinicians.
- Capable of establishing strong working relationships across the organization to define and solve problems that will benefit the whole.
- Ability to manage multiple projects simultaneously and flexibly manage priorities.
- Experience debugging complex systems is a plus.
- Experience with mechanistic ADMET models, PK/PD and PBPK modeling is a plus.
Notice to Search Firms/Third-Party Recruitment Agencies (Recruiters)
The Human Resources team manages the recruitment and employment process for Treeline Biosciences. To protect the interests of all parties involved, Treeline Biosciences will only accept resumes from a recruiter if an executed search agreement directed to the particular position or positions is in place at the start of the recruitment effort. Unsolicited resumes sent to Treeline Biosciences from recruiters do not constitute any type of relationship between the recruiter and Treeline Biosciences and do not obligate Treeline Biosciences in any way to pay fees should we hire from those resumes. Recruiters are requested not to contact or present candidates directly to our hiring manager or employees.