UKBB-Data

1. Dataset description

The UK Biobank is a large-scale, population-based cohort study involving over 500,000 UK residents aged 40–69 years. Recruitment took place between 2006 and 2010, during which participants completed extensive health questionnaires, underwent detailed physical assessments, and provided biological samples. The study tracked health-related events through data linkage to hospital admission records and mortality registries. In 2009, the UK Biobank introduced ophthalmic examinations to further expand the scope of the collected data[1].

Homepage of the dataset： https://www.ukbiobank.ac.uk/

2. Our current dataset (continuously updated)

The data we currently possess mainly includes: Retinal images, and health-realted data.

Dataset path: /scratch/users/nus/hongyu.h

1. Retianl images

Dataset path: /scratch/users/nus/hongyu.h/SERI_UKbiobank/1_Fundus_Images

a. #NUMBER

  UKBB_FP_01 : 46514  
  UKBB_FP_02 : 40826  
  UKBB_FP_03 : 38786  
  UKBB_FP_04 : 49705  
  UKB_new_2024: 1169 + 949 + 1154 + 1 +923 = 4196  
  Total count of retinal images：180027  

b. INSTANCE

Interpretation of the naming format： Participants ID_Data field_Instances_Array

Detailed information is available on the official website.

1017168_21016_2_1.png

1021021_21015_2_1.png

2.Health-related data

Dataset path: /scratch/users/nus/hongyu.h/SERI_UKbiobank/2_Phenptype

The raw data we have is stored in the "/0_Header_Tab" folder, and the other folders contain the organized data required for other projects.

In the "/0_Header_Tab" folder, we have header.txt file and its corresponding xxx.tab file. The .tab file serves as the main data file, and the header file is created by extracting the header from the main data file.

Therefore, if you want to find the data needed for your project, you can search for the field ID on the UK Biobank website and then check if the header files contain the field ID.

---

Example of extracting ethnic data

If we want to extract ethnic data, we need to go to the UKBB website and find the field ID for ethnic data.

It can be found that the field ID for ethnic data is 21000, and detailed information about the ethnic data can also be obtained.

Next, let's find the location of the ethnic data within the header file.

grep -n '21000' *.txt

We can obtain the search result

D3_ukb41252_tab_8330_header.txt:4737:  4737  f.20160.0.0
D3_ukb41252_tab_8330_header.txt:4738:  4738	f.20160.1.0  
D3_ukb41252_tab_8330_header.txt:4739:  4739	f.20160.2.0  
D3_ukb41252_tab_8330_header.txt:4740:  4740	f.20160.3.0

According to the ethnic data detailed information, we select 'f.20160.0.0.', which is located at row 4737 in D3_ukb41252_tab_8330_header.txt.

Then, we can extract the ethnic data from the corresponding tab file.

cut -f 1,4737 /scratch/users/nus/hongyu.h/SERI_UKBiobank/2_Phenotype/ukb41252.tab > xxx/ethnic.txt

Through the simple steps mentioned above, we can locate and extract the necessary data for the project.

---

3.Our project data

I've created a "3_Project" folder to help keep our project data organized. Feel free to create your own subfolders within it to store any data you've prepared for your specific needs.

To make things easier, I've also set up a "basic" folder that has essential data like gender and ethnic, since most of our projects rely on this info. Having it all in one place should save us some time and effort.

Let me know if you have any thoughts or suggestions!

Reference

[1] Zhu, Zhuoting, et al. "Retinal age gap as a predictive biomarker for mortality risk." British Journal of Ophthalmology 107.4 (2023): 547-554.