Login
 Forum
 
 
Thesis topic proposal
 
Csaba Kerepesi
Age prediction based on random sequencing data

THESIS TOPIC PROPOSAL

Institute: Eötvös Loránd University, Budapest
mathematics and computing
Doctoral School of Mathematics

Thesis supervisor: Csaba Kerepesi
Location of studies (in Hungarian): HUN-REN SZTAKI
Abbreviation of location of studies: H-REN


Description of the research topic:

Aging clocks are machine learning models that can predict age and presumably also give a good estimation of biological age. They can apply to potential longevity and rejuvenation treatments. Currently, the most popular and useful aging clocks are the array-based epigenetic clocks. They use a fixed number of features (typically 850,000), where the features are methylation levels of certain positions of the DNA. Another type of method is the bisulfite sequencing-based epigenetic clock which would be theoretically more accurate than the array-based ones as they can measure a higher number of positions. However, the randomness of the DNA sequencing prevents applicability. In this measurement technique the features are randomly selected for every sample resulting in inaccurate predictions for the test set due to the high number of missing features of the model. The task of the candidate is comprehensive testing of different mathematical modeling and machine learning methods (e.g. missing value imputation techniques, simulations, probabilistic methods, and the “intersection clock” recently developed by the supervisor https://doi.org/10.1111/acel.13922) as well as developing new ones that can be an efficient solution for the problem. Numerous available datasets (that can be interesting e.g. for rejuvenation research) are measured by the bisulfite sequencing technique. Applying the newly developed methods, the candidate can make important discoveries. Another research direction is the generalization of the methods to different domains where the feature set has a randomity.

Required language skills: English
Further requirements: 
Excellent mathematics and informatics knowledge with interest in the aging and rejuvenation research.

Number of students who can be accepted: 1

Deadline for application: 2024-05-31


2024. IV. 17.
ODT ülés
Az ODT következő ülésére 2024. június 14-én, pénteken 10.00 órakor kerül sor a Semmelweis Egyetem Szenátusi termében (Bp. Üllői út 26. I. emelet).

 
All rights reserved © 2007, Hungarian Doctoral Council. Doctoral Council registration number at commissioner for data protection: 02003/0001. Program version: 2.2358 ( 2017. X. 31. )