M.T. Patrick

A Review of Recent Advancement in Integrating Omics Data with Literature Mining towards Biomedical Discoveries

In the past decade, the volume of “omics” data generated by the different high-throughput technologies has expanded exponentially. The managing, storing, and analyzing of this big data have been a great challenge for the researchers, especially when moving towards the goal of generating testable data-driven hypotheses, which has been the promise of the high-throughput experimental techniques. Different bioinformatics approaches have been developed to streamline the downstream analyzes by providing independent information to interpret and provide biological inference. Text mining (also known as literature mining) is one of the commonly used approaches for automated generation of biological knowledge from the huge number of published articles. In this review paper, we discuss the recent advancement in approaches that integrate results from omics data and information generated from text mining approaches to uncover novel biomedical information.

Machine learning workflow to enhance predictions of Adverse Drug Reactions (ADRs) through drug-gene interactions: application to drugs for cutaneous diseases

Adverse drug reactions (ADRs) pose critical public health issues, affecting over 6% of hospitalized patients. While knowledge of potential drug-drug interactions (DDI) is necessary to prevent ADR, the rapid pace of drug discovery makes it challenging to maintain a strong insight into DDIs. In this study, we present a novel literature-mining framework for enhancing the predictions of DDIs and ADR types by integrating drug-gene interactions (DGIs). The ADR types were adapted from a DDI corpus, including i) adverse effect; ii) effect at molecular level; iii) effect related to pharmacokinetics; and iv) DDIs without known ADRs. By using random forest classifier our approach achieves an F-score of 0.87 across the ADRs classification using only the DDI features. We then enhanced the performance of the classifier by including DGIs (F-score = 0.90), and applied the classification model trained with the DDI corpus to identify the drugs that might interact with the drugs for cutaneous diseases. We successfully predict previously known ADRs for drugs prescribed to cutaneous diseases, and are also able to identify promising new ADRs.

Photosensitivity and type I IFN responses in cutaneous lupus are driven by epidermal-derived interferon kappa

Objective Skin inflammation and photosensitivity are common in patients with cutaneous lupus erythematosus (CLE) and systemic lupus erythematosus (SLE), yet little is known about the mechanisms that regulate these traits. Here we investigate the role of interferon kappa (IFN-κ) in regulation of type I interferon (IFN) and photosensitive responses and examine its dysregulation in lupus skin. Methods mRNA expression of type I IFN genes was analysed from microarray data of CLE lesions and healthy control skin. Similar expression in cultured primary keratinocytes, fibroblasts and endothelial cells was analysed via RNA-seq. IFNK knock-out (KO) keratinocytes were generated using CRISPR/Cas9. Keratinocytes stably overexpressing IFN-κ were created via G418 selection of transfected cells. IFN responses were assessed via phosphorylation of STAT1 and STAT2 and qRT-PCR for IFN-regulated genes. Ultraviolet B-mediated apoptosis was analysed via TUNEL staining. In vivo protein expression was assessed via immunofluorescent staining of normal and CLE lesional skin. Results IFNK is one of two type I IFNs significantly increased (1.5-fold change, false discovery rate (FDR) q<0.001) in lesional CLE skin. Gene ontology (GO) analysis showed that type I IFN responses were enriched (FDR=6.8×10−04) in keratinocytes not in fibroblast and endothelial cells, and this epithelial-derived IFN-κ is responsible for maintaining baseline type I IFN responses in healthy skin. Increased levels of IFN-κ, such as seen in SLE, amplify and accelerate responsiveness of epithelia to IFN-α and increase keratinocyte sensitivity to UV irradiation. Notably, KO of IFN-κ or inhibition of IFN signalling with baricitinib abrogates UVB-induced apoptosis. Conclusion Collectively, our data identify IFN-κ as a critical IFN in CLE pathology via promotion of enhanced IFN responses and photosensitivity. IFN-κ is a potential novel target for UVB prophylaxis and CLE-directed therapy.

Genetic signature to provide robust risk assessment of psoriatic arthritis development in psoriasis patients

Psoriatic arthritis (PsA) is a complex chronic musculoskeletal condition that occurs in ~30% of psoriasis patients. Currently, no systematic strategy is available that utilizes the differences in genetic architecture between PsA and cutaneous-only psoriasis (PsC) to assess PsA risk before symptoms appear. Here, we introduce a computational pipeline for predicting PsA among psoriasis patients using data from six cohorts with >7000 genotyped PsA and PsC patients. We identify 9 new loci for psoriasis or its subtypes and achieve 0.82 area under the receiver operator curve in distinguishing PsA vs. PsC when using 200 genetic markers. Among the top 5% of our PsA prediction we achieve >90% precision with 100% specificity and 16% recall for predicting PsA among psoriatic patients, using conditional inference forest or shrinkage discriminant analysis. Combining statistical and machine-learning techniques, we show that the underlying genetic differences between psoriasis subtypes can be used for individualized subtype risk assessment.Approximately 30% of psoriasis patients develop psoriatic arthritis (PsA) and early diagnosis is crucial for the management of PsA. Here, Patrick et al. develop a computational pipeline involving statistical and machine-learning methods that can assess the risk of progression to PsA based on genetic markers.

Drug repurposing prediction for immune-mediated cutaneous diseases using a word-embedding based machine learning approach.

Immune-mediated diseases affect more than 20% of the population, and many autoimmune diseases affect the skin. Drug repurposing (or repositioning) is a cost-effective approach for finding drugs that can be used to treat diseases for which they are currently not prescribed. We implemented an efficient bioinformatics approach using word embedding to summarize drug information from more than 20 million articles and applied machine learning to model the drug-disease relationship. We trained our drug repurposing approach separately on nine cutaneous diseases (including psoriasis, atopic dermatitis, and alopecia areata) and eight other immune-mediated diseases and obtained a mean area under the receiver operating characteristic of 0.93 in cross-validation. Focusing in particular on psoriasis, a chronic inflammatory condition of skin that affects more than 100 million people worldwide, we were able to confirm drugs that are known to be effective for psoriasis and to identify potential candidates used to treat other diseases. Furthermore, the targets of drug candidates predicted by our approach were significantly enriched among genes differentially expressed in psoriatic lesional skin from a large-scale RNA sequencing cohort. Although our algorithm cannot be used to determine clinical efficacy, our work provides an approach for suggesting drugs for repurposing to immune-mediated cutaneous diseases.

Research Techniques Made Simple: Using Genome-Wide Association Studies to Understand Complex Cutaneous Disorders

Complex cutaneous disorders result from the combined effect of many different genes and environmental factors, with individual genetic variants often having only a modest effect on disease risk. The ability to examine large numbers of samples is required for correlating genetic variants with diseases/traits. Technological advances in high-throughput genotyping, along with mapping of the human genome and its associated inter-individual variation, have allowed genetic variants to be analyzed at high density in large case-control cohorts for many diseases, including several major skin diseases. These genome-wide association studies focus on showing differences in the frequencies of variants between case and control groups, rather than co-transmission of a variant and disease through a family, as is done in linkage studies. In this review, we provide overall guidance for genome-wide association study analysis and interpreting the results. Additionally, we discuss challenges and future directions for genome-wide association studies, focusing on translation of findings to provide biological and clinical implications for dermatology.