applying
different time-series interpolation techniques on real process datasets

Objective:

In the last 10 years, new molecular biology tools like Next Generation Sequencing (NGS) became available to unravel the microbial composition and activity of microbial communities. This offers the potential of providing information on both the presence and the response of the microorganisms towards their environment, allowing a comprehensive evaluation of the water quality in wastewater treatment plants.

The meta’omics datasets which are generated by NGS technology are big data in nature and provide an opportunity for tapping into new knowledge potential by using data-driven models. However, fitting reliable models using meta’omics dataset can be a challenging task due to high dimensionality, limited amount of samples, high amount of unknowns in the dataset, and great disparity in the sampling frequencies.

To circumvent problems caused by missing data points, different time-series interpolation techniques exist. For this internship project, a comprehensive performance evaluation of different time-series interpolation techniques will be conducted on a real-life data set to model water microbiome diversity as a measure of different process parameters in a saline wastewater treatment plant in the Netherlands.

Requirements:

• EU citizen or international student registered at a Dutch university
• Specialized in Data Science, Machine Learning, Engineering, Artificial Intelligence, Bio-informatics or bio-statistics
• Experience with R or Python is a requirement
• Experience with GitHub is a requirement
• Fluency in English
• Interest in water systems and in water microbiomes is an advantage

Contact details:
Please send your CV (max two pages) and motivation letter (max 1 page) using the form below. Do not hesitate to contact me (asala.mahajna@wetsus.nl) in case you have any further questions or need more information.