Caring for the world, one person at a time has inspired and united the people of Johnson & Johnson for over 130 years. We embrace research and science -- bringing innovative ideas, products and services to advance the health and well-being of people. With $76.5 billion in 2017 sales, Johnson & Johnson is the world's most comprehensive and broadly-based manufacturer of health care products, as well as a provider of related services, for the consumer, pharmaceutical, and medical devices markets. There are more than 250 Johnson & Johnson operating companies employing over 125,000 people and with products touching the lives of over a billion people every day, throughout the world. If you have the talent and desire to touch the world, Johnson & Johnson has the career opportunities to help make it happen. Thriving on a diverse company culture, celebrating the uniqueness of our employees and committed to inclusion. Proud to be an equal opportunity employer.
Developing innovative therapeutics to treat diseases like Alzheimer’s disease, various types of cancers and infectious diseases like Hepatitis B, influenza is our passion. In this endeavor, we are seeking to recruit new talent for the comprehensive analyses of high-dimensional datasets using state-of-the-art data science methods applied to drug discovery programs. The position is opened at Spring House (PA), a headquarters of Janssen Research & Development. We significantly increased our investment into the workforce for data analysis pipelines with the emphasis in current cutting-edge technologies to support future Artificial Intelligence-driven drug design and discovery.
Janssen Research & Development L.L.C., is looking for the 2-year postdoctoral fellow to support drug design and discovery projects using machine learning approaches with emphasis on transfer learning and deep learning. Deep learning techniques have already shown promise for small molecule projects in Janssen, yet most of those models require a significant amount of data, while many of the ADMETox-related pipelines generate significantly smaller datasets that require transfer learning to integrate them successfully into the predictive pipelines.
This position will support small molecule design and optimization using machine learning techniques by integrating millions of data points coming from heterogeneous data sources: chemical structure, microscopy images, and various omics experiments. The primary goal is the improvement of the predictive pipelines to increase safety and efficacy of the drug candidates and decrease the time needed to progress hit compound to lead compound to compound in clinical trials.
Main responsibilities would include the development of predictive models and their testing in real projects that would require interaction with chemists, biologists, and data scientist and further model optimization if needed. We are looking for candidates with a track record in deep learning and preferably experience working with chemical or biological data.
- Support drug design and development using cutting edge machine learning techniques;
- Design and development of the transfer learning pipelines for small and medium size datasets (SMSDS) from scratch (Keras, TensorFlow, etc.) or adaptation of the open source code if available;
- Finding the beneficial interplay of the transfer learning on large datasets (millions of compounds by thousands of biological end-point types) to small and medium size datasets reflecting ADMET assays (thousands of end-points per assay);
- Building machine learning models that effectively learn from heterogenous multi-modal data, e.g. modeling biological effects of compounds on the basis of chemical structure and gene expression, high content image descriptors or other omics data
- Application of the developed pipelines in drug design and development project with actionable conclusions;
- Integration of the pipelines into internal expert system together with end-users (chemists, biologists, data scientists) and IT support: to promote transparency, traceability and visual component of the developed technique.
- Contributing to the scientific weight of the department by authoring peer-reviewed papers and presenting at relevant conferences.
Preferably PhD in Computer Science, Data Science, Computational chemistry, Bioinformatics or a related quantitative field, preferably with experience of interaction with research chemists or biologists in an academic or industrial setting or equivalent experience
Advanced programming skills to enable the development of functional prototypes
Experience with Deep Learning machine learning frameworks, like PyTorch, Keras, Tensorflow or alike
Excellent communication, reporting and team interaction skills, self-motivation, proactivity and the ability to work independently
A passion for hands-on science and delivering high quality results in the lab.
Good communication, organizing, and planning skills and the ability to take leadership for and drive decision making in a research project.
Ability to develop and deliver a presentation on technical data with strong business impact.
Creative mind that is able to see things from different perspectives and come up with innovative solutions to complex problems. Ability to introduce best practices from previous work experiences into the new team.
Desire for continuous learning and the ability to identify, evaluate and implement emerging areas of science.
Johnson & Johnson is an Affirmative Action and Equal Opportunity Employer. All qualified applicants will receive consideration for employment without regard to race, color, religion, sex, sexual orientation, gender identity, age, national origin, or protected veteran status and will not be discriminated against on the basis of disability.
United States-Pennsylvania-Spring House-
Janssen Research & Development, LLC (6084)