PhD position: Deep learning as a computational modelling technique for genomics
Supervisor: Prof. Ivo Provazník
As a data-driven science, genomics utilizes machine learning to search for dependencies in data and hypothesize novel biological phenomena. The need for extraction of new insights from the exponentially increasing volume of genomics data requires more expressive machine learning models. Deep learning is becoming the method of choice for many genomics modelling tasks such as predicting the impact of genetic variation on gene regulatory mechanisms.
The main aim of the project is to design novel tools for genomic data partitioning and prediction, fitting parameters and choosing hyperparameters for optimal training of deep neural networks. The tools will be used to discover local patterns and longe-range dependencies in sequential data and modelling transcription factor binding sites and spacing. The project requires deep research in the field of existing machine learning applications for the analysis of genome sequencing data sets, including the annotation of sequence elements and epigenetic, proteomic or metabolomic data. Supervised, semi-supervised and unsupervised machine learning methods, as well as of generative and discriminative modelling approaches will be considered. The capacity of deep learning models to identify transcription factors binding sites from DNA sequences will be investigated based on searching dependencies in the data.
For more information about this topic please contact Prof. Ivo Provazník – email@example.com.
- Kopp W, et al. Deep learning for genomics using Janggu. Nat Commun 11, 3488 (2020)
- Eraslan G, et al. Deep learning: new computational modelling techniques for genomics. Nat Rev Genet 20, 389-403 (2019)
Who we are seeking:
- Deep interest in artifficial intelligence approaches for medical applications
- Deep interest in genomics
- A sound knowledge of programming languages (eg. Python, R)
- A relevant degree with appropriate engineering and/or IT knowledge, transferable to the scientific environment
- English communication skills
What we offer:
- Our core objective is to provide the doctoral students with a supportive and highly scientific work environment that fosters collaboration
- The doctoral students complete 3-6 months of internships at partner universities abroad
- The Department provides doctoral students with a scholarship beyond the state scholarship in the form of a supplementary stipend or salary when participating in a grant project
You are invited to apply through university e-application at:
Deadline for application: May 15, 2021
Additional documents to be submitted: