Missing Data Imputation in Biology

Missing data is a common issue in biological research, often arising from experimental errors, equipment malfunctions, or sample degradation. This project aims to evaluate various imputation methods specifically for missing data in biological contexts. The student can also choose to examine the scalability of known methods or to develop new techniques that are scaleable.

Goal

• To test and compare various imputation methods on biological datasets that have non-randomly missing values in terms of effectiveness and computation times. This will provide insights into which methods are most effective and under what conditions.

Learning outcome

• Use state-of-the-art imputation techniques to handle missing data for biological data

Qualifications

• Proficient in either Python or R

Supervisors

• Thu Thi Nguyen
• Michael Riegler
• Pål Halvorsen

Collaboration partners

• Marcin Wojewodzic from Norwegian Cancer Registry