Journal: Frontiers in Genetics

Article Title: Impute.me: An Open-Source, Non-profit Tool for Using Data From Direct-to-Consumer Genetic Testing to Calculate and Interpret Polygenic Risk Scores

doi: 10.3389/fgene.2020.00578

Figure Lengend Snippet: Basic pipeline setup from the user point of view. On upload of a genome, data are checked according quality control (QC) parameters that have been developed to handle most types of microarray-based consumer genetics data. The genome is then imputed using 1000 Genomes as reference ( left ). The imputed data are then further subjected to automated analysis scripts from 15 different modules, most of which are based on polygenic risk score calculations. The calculations include 1,859 traits from genome-wide association studies (GWASs) and 634 traits from the UK Biobank, as well as customized modules for height, and drug response. Most polygenic risk scores use GWAS significant single-nucleotide polymorphisms (SNPs) out of necessity, although 20 major diseases are based on LDpred all-SNP scores ( center ). A user can then browse their scores in relation to the population, shown together with a chart displaying how much variability is explained ( right ).

Article Snippet: It also expands the SNP overlap between microarray types from the major vendors, such as 23andMe, MyHeritage, and Ancestry.com .

Techniques: Control, Microarray, GWAS

Journal: Frontiers in Genetics

Article Title: Impute.me: An Open-Source, Non-profit Tool for Using Data From Direct-to-Consumer Genetic Testing to Calculate and Interpret Polygenic Risk Scores

doi: 10.3389/fgene.2020.00578

Figure Lengend Snippet: Theoretical background of the analysis pipeline. (A) Clinical genetics currently concern high-effect DNA variants that often can only be sequenced ( red ). Additionally, high-effect variants such as APOE4 and a small subset of BRCA1 and BRCA2 pathogenic variants are possible to measure using microarray ( blue includes several other variants not shown in plot, e.g., Parkinson’s variants). There may be an untapped potential for valuable clinical information in polygenic risk scores (PRSs) for common disease ( green ), for example, for type 2 diabetes (T2D), coronary artery disease (CAD), or statin response ( ; ; ). It is a primary aim of the impute.me project to make this potential available more broadly, balancing the practice of relying on individual genome-wide association study (GWAS) single-nucleotide polymorphisms (SNPs) and/or reporting of SNP genotypes ( pink ). (B) The secondary aim is to provide genetic scores in a relevant context, exemplified in the precision medicine module showing the so-called health-context tree. This tree consists of all entries from the international classification of disease [ International Classification of Diseases , 10th Revision (ICD-10)], linked to all genetic studies. It allows browsing of PRSs in a relevant context. In the example shown, the tree is open on the psychiatry chapter, showing PRSs for schizophrenia (F20), unipolar depression (F32), and bipolar depression (F31). Although these scores have little predictive relevance for a healthy individual, they may be useful in the context of psychiatric evaluation, particularly in the case of more extreme scores.

Article Snippet: It also expands the SNP overlap between microarray types from the major vendors, such as 23andMe, MyHeritage, and Ancestry.com .

Techniques: Microarray, GWAS, Clinical Proteomics