Data Science Postdoc Position in Machine Learning assisted identification of biological pesticide replacements for sustainable agriculture is available at DTU Bioengineering, Technical University of Denmark, Denmark. (Data Scientist Job)
Deadline to Apply
May 17, 2021 (23: 59 GMT +1)
|No. of Position(s)||1|
|Research Area||– Data Science|
– Computational Biology,
– Applied Mathematics,
– Computer Science
|Salary||According to standard norms|
|Workplace||Computer Aided Biotechnology – DTU Bioengineering|
Technical University of Denmark,
|Contract Period||2.5 Year|
|Starting date||Jul 01, 2021|
- PhD degree (or equivalent) in Data Science or a related field (Bioinformatics, Computational Biology, Statistics, Applied Mathematics, Computer Science etc.).
Furthermore, you must have the following qualifications and skills:
- Proficient in using the Python Data Science stack (NumPy, SciPy, Pandas, Scikit-learn, etc.) and visualization libraries (altair, plotly, matlotlib, etc.).
- Experience with Deep Learning and at least one corresponding framework (PyTorch, Tensorflow, etc.).
- Comfortable in modeling complex data by communicating with domain experts.
- Proficient with basic back-end development using relational databases, ORMs etc. preferably in Python.
- Good coding practices like test-driven development, version control with git, code reviews etc.
Preferred additional qualifications include:
- Real-world project experience with Active Learning and/or Bayesian Optimization for experimental design is a plus.
- Experience with image processing and analysis using skimage, opencv and/or CNNs is a plus.
- Experience with Airflow (or similar) for managing data pipelines and ETL workflows is a plus.
- Experience with cloud computing, Docker, Kubernetes is a plus.
- Experience with using NoSQL/graph databases is a plus.
- Bioinformatics experience with microbial genome sequence data processing (assembly and annotation) is a plus.
Responsibilities/ Job Description
- As part of a multi-disciplinary team, your primary responsibility will be to develop the Machine Learning based experimental design pipeline to significantly reduce the vast search space of available microorganisms through iterative and model-guided experimentation using Active Learning, Bayesian Optimization, and other suitable approaches.
Tasks you will carry out for this purpose include:
- Develop and validate predictive Machine Learning models based on iteratively generated biological data.
- Utilize and communicate model predictions to determine next iteration of most informative experiments.
- Develop data management solutions and guide experimentalists in systematic and automated data capture.
- Support other team members in basic data processing tasks such as time series analysis and image processing.
How to Apply?
To apply, please open the link “Apply online”, fill out the online application form, and attach all your materials in English in one PDF file.
Applications must be submitted as one PDF file containing all materials to be given consideration. The file must include:
- Application (cover letter)
- Academic Diplomas (MSc/PhD)
- List of publications
Your complete online application must be submitted no later than 17 May 2021 (Danish time) – but please submit early as we will start a preliminary review and interview process.
About Research project
The postdoc position is part of the Smarter AgroBiological Screening (SABS) project, which will develop new biological antifungal agents for crop protection, to enable the reduction of chemical fungicides usage. The project is an industry-university collaboration involving a large team at DTU Bioengineering, to build-up a set of high-throughput automated assays to screen DTU’s collection of microorganisms and advance most promising strains towards in-planta testing. Artificial intelligence methodologies will be used to find predictive patterns in the collected data and enable most effective selection and optimization process. The SABS project is funded by the Innovation Fund Denmark and is located at DTU Bioengineering – Department of Biotechnology and Biomedicine.
Presently, the state-of-the-art approach of brute force screening of biological agents is simply too costly and time consuming, making it challenging to compete with existing chemical-based pesticides. So, innovative ideas and methodologies are needed to navigate this vast search space effectively. From a Machine Learning perspective, this problem boils down to a scenario of abundant unlabeled data (vast number of microorganisms are available for screening) that are costly to label experimentally (“Is this organism a crop protective agent?”). For this scenario, approaches such as Active Learning and Bayesian Optimization can be utilized to facilitate a cost-effective generation of data points.
All interested candidates irrespective of age, gender, disability, race, religion or ethnic background are encouraged to apply.
Associate Professor Nikolaus Sonnenschein
Data Science Postdoc in Machine Learning assisted identification of biological pesticide replacements for sustainable agriculture
Post expires at 8:59am on Tuesday May 18th, 2021 (GMT+9)