PINES (Phenotype-Informed Noncoding Element Scoring)
PINES (Phenotype-Informed Noncoding Element Scoring)

Functional characterization of the noncoding genome is essential for the biological understanding of gene regulation and disease. Here, we introduce the computational framework PINES (Phenotype-Informed Noncoding Element Scoring) which evaluates the functional impact of noncoding variants in an unsupervised setting by integrating epigenetic annotations from diverse sources in a phenotype-dependent manner. A unique feature of PINES is that analyses may be customized towards genomic annotations of the highest relevance to phenotypes of interest. We show that PINES identifies functional noncoding variation more accurately than methods that do not use phenotype-weighted knowledge. To demonstrate its versatility, we show how PINES can be applied to systematically distinguish variants in enhancer regions from background variation, to delineate high-penetrance noncoding variants leading to Mendelian phenotypes, or to pinpoint epigenetic signals underlying noncoding alleles at GWAS loci. We further show how PINES can help interpret the results of fine-mapping approaches and predict novel causal alleles from GWAS on Parkinson's and inflammatory bowel disease. Its flexibility and ease of use through a dedicated web portal establish PINES as a powerful new in silico method to prioritize and assign functions to noncoding genetic variants.

The manuscript can be found here. This online interface can be used to score up to 30,000 noncoding variants per job. Visit our GitHub page to download the full source code (under the GNU GPL) and annotation data. E-mail us with questions or if you have any technical problems using PINES.


Variants to score
(one rs ID per line)
input (optional)
Tissue:    and manual weighting constant:  
Variants for enrichment
weighting (one rs ID per line):  
Email address

Merck Harvard Medical School Brigham and Women's Hospital