Towards precision medicine: interrogating the human genome to identify drug pathways associated with potentially functional, population-differentiated polymorphisms.

Drug response variations amongst different individuals/populations are influenced by several factors including allele frequency differences of single nucleotide polymorphisms (SNPs) that functionally affect drug-response genes. Here, we aim to identify drugs that potentially exhibit population differences in response using SNP data mining and analytics. Ninety-one pairwise-comparisons of >22,000,000 SNPs from the 1000 Genomes Project, across 14 different populations, were performed to identify 'population-differentiated' SNPs (pdSNPs). Potentially-functional pdSNPs (pf-pdSNPs) were then selected, mapped into genes, and integrated with drug-gene databases to identify 'population-differentiated' drugs enriched with genes carrying pf-pdSNPs. 1191 clinically-approved drugs were found to be significantly enriched (Z > 2.58) with genes carrying SNPs that were differentiated in one or more population-pair comparisons. Thirteen drugs were found to be enriched with such differentiated genes across all 91 population-pairs. Notably, 82% of drugs, which were previously reported in the literature to exhibit population differences in response were also found by this method to contain a significant enrichment of population specific differentiated SNPs. Furthermore, drugs with genetic testing labels, or those suspected to cause adverse reactions, contained a significantly larger number (P < 0.01) of population-pairs with enriched pf-pdSNPs compared with those without these labels. This pioneering effort at harnessing big-data pharmacogenomics to identify 'population differentiated' drugs could help to facilitate data-driven decision-making for a more personalized medicine.

Architecture of population-differentiated polymorphisms in the human genome.

Population variation in disease and other phenotype are partly attributed to single nucleotide polymorphisms (SNPs) in the human genome. Due to selection pressure, two individuals from the same ancestral population have more genetic similarity compared to individuals from further geographic regions. Here, we elucidated the genomic population differentiation pattern, by interrogating >22,000,000 SNPs. Majority of population-differentiated (pd) SNPs (~95%), including the potentially functional (pf) (~84%) subset reside in non-genic regions, compared to the proportion of all SNPs (58%) found in non-genic regions. This suggests that differences between populations are more likely due to differences in gene regulation rather than protein function. Actin Cytoskeleton, Axonal Guidance and Protein Kinase A signaling pathways are enriched with genes carrying at least three pdSNPs (enriched pdGenes), while Antigen Presentation, Hepatic Fibrosis and Huntington Disease Signalling pathways are over-represented by enriched pf-pdGenes. An inverse correlation between chromosome size and the proportion of pd-/pf-pdSNPs was observed. Smaller chromosomes have relatively more of such SNPs including genes carrying these SNPs. Genes associated with common diseases and enriched with these pd-/pfpdSNPs are localized to 11 different chromosomes, with immune-related disease pd/pf-pdGenes mainly residing in chromosome 6 while neurological disease pd/pf-pdGenes residing in smaller chromosomes including chromosome 21/22. The associated diseases were reported to show population differences in incidence, severity and/or etiology. In summary, this study highlights the non-sporadic nature of population differentiation footprint in the human genome, which can potentially lead to the identification of genomic regions that play roles in the manifestation of phenotypic differences, including in disease predisposition and drug response.

Comprehensive review of Hepatitis B Virus-associated hepatocellular carcinoma research through text mining and big data analytics.

PubMed was text mined to glean insights into the role of Hepatitis B virus (HBV) in hepatocellular carcinoma (HCC) from the massive number of publications (9249) available to date. Reports from ∼70 countries identified >1300 human genes associated with either the Core, Surface or X gene in HBV-associated HCC. One hundred and forty-three of these host genes, which can potentially yield 1180 biomolecular interactions, each were reported in at least three different publications to be associated with the same HBV. These 143 genes function in 137 pathways, involved mainly in the cell cycle, apoptosis, inflammation and signalling. Fourteen of these molecules, primarily transcriptional regulators or kinases, play roles in several pathways pertinent to the hallmarks of cancers. 'Chronic' was the most frequent word used across the 9249 abstracts. A key event in chronic HBV infection is the integration of HBV into the host genome. The advent of cost-effective, next-generation sequencing technology facilitated the employment of big-data analytics comprehensively to characterize HBV-host integration within HCC patients. A total of 5331 integration events were reported across seven publications, with most of these integrations observed between the Core/X gene and the introns of genes. Nearly one-quarter of the intergenic integrations are within repeats, especially long interspersed nuclear elements (LINE) repeats. Integrations within 13 genes were each reported by at least three different studies. The human gene with the most HBV integrations observed is the TERT gene where a total of 224 integrations, primarily at its promoter and within the tumour tissue, were reported by six of seven publications. This unique review, which employs state-of-the-art text-mining and data-analytics tools, represents the most complete, systematic and comprehensive review of nearly all the publications associated with HBV-associated HCC research. It provides important resources to either focus future research or develop therapeutic strategies to target key molecules reported to play important roles in key pathways of HCC, through the systematic analyses of the commonly reported molecules associated with the various HBV genes in HCC, including information about the interactions amongst these commonly reported molecules, the pathways in which they reside as well as detailed information regarding the viral and host genes associated with HBV integration in HCC patients. Hence this review, which highlights pathways and key human genes associated with HBV in HCC, may facilitate the deeper elucidation of the role of HBV in hepato-carcinogenesis, potentially leading to timely intervention against this deadly disease.

Architecture of polymorphisms in the human genome reveals functionally important and positively selected variants in immune response and drug transporter genes.

BACKGROUND: Genetic polymorphisms can contribute to phenotypic differences amongst individuals, including disease risk and drug response. Characterization of genetic polymorphisms that modulate gene expression and/or protein function may facilitate the identification of the causal variants. Here, we present the architecture of genetic polymorphisms in the human genome focusing on those predicted to be potentially functional/under natural selection and the pathways that they reside. RESULTS: In the human genome, polymorphisms that directly affect protein sequences and potentially affect function are the most constrained variants with the lowest single-nucleotide variant (SNV) density, least population differentiation and most significant enrichment of rare alleles. SNVs which potentially alter various regulatory sites, e.g. splicing regulatory elements, are also generally under negative selection. Interestingly, genes that regulate the expression of transcription/splicing factors and histones are conserved as a higher proportion of these genes is non-polymorphic, contain ultra-conserved elements (UCEs) and/or has no non-synonymous SNVs (nsSNVs)/coding INDELs. On the other hand, major histocompatibility complex (MHC) genes are the most polymorphic with SNVs potentially affecting the binding of transcription/splicing factors and microRNAs (miRNA) exhibiting recent positive selection (RPS). The drug transporter genes carry the most number of potentially deleterious nsSNVs and exhibit signatures of RPS and/or population differentiation. These observations suggest that genes that interact with the environment are highly polymorphic and targeted by RPS. CONCLUSIONS: In conclusion, selective constraints are observed in coding regions, master regulator genes, and potentially functional SNVs. In contrast, genes that modulate response to the environment are highly polymorphic and under positive selection.