Batch 1+2 harmonized data

The harmonised and QCd Olink dataset combines batches 1 and 2 (n=4,415) of 5.4k proteins. A detailed file was circulated with the data release, with the necessary details on how the merging was done. The file can also be found in FinnGen Teams/Data release announcements/Shared/Proteomics with the name "DATA_RELEASE_NOTE_FG3_CrossBatch_Harmonised_March2026".

You can find the data in

library-red/omics/proteomics/olink_genewiz_batch1_and_2_harmonized_QCd_2026_16_03

pQTL analysis

The data and full documentation are available at: /finngen/library-green/omics/proteomics/olink_genewiz_batch1_and_2_QCd_2026_01_05/

In addition to the fine-mapped results, we have released the pQTL summary statistics and LDSC-formatted pre-munge summary statistics. The release note also details a formal comparison of the FG3 fine-mapping output against two benchmark cohorts: FinnGen Olink Batch 01 and UKB-PPP.

Upstream proteomics normalization, covariate adjustment, and kinship filtering remain as described in the previous release. For details on upstream processing, please consult the March 2026 (v1.1) documentation: DATA_RELEASE_NOTE_FG3_CrossBatch_Harmonised_v01_1.pdf in /finngen/library-red/omics/proteomics/olink_genewiz_batch1_and_2_harmonized_QCd_2026_16_03