Prostate cancer is a heritable disease with ancestry-biased incidence and mortality. Polygenic risk scores (PRSs) offer promising advancements in predicting disease risk, including prostate cancer. While their accuracy continues to improve, research aimed at enhancing their effectiveness within African and Asian populations remains key for equitable use. Recent algorithmic developments for PRS derivation have resulted in improved pan-ancestral risk prediction for several diseases. In this study, we benchmark the predictive power of six widely used PRS derivation algorithms, including four of which adjust for ancestry, against prostate cancer cases and controls from the UK Biobank and All of Us cohorts. We find modest improvement in discriminatory ability when compared with a simple method that prioritizes variants, clumping, and published polygenic risk scores. Our findings underscore the importance of improving upon risk prediction algorithms and the sampling of diverse cohorts.

In this work, we proposed a DDI detection method based on molecular structures using graph convolutional networks and deep sets. We proposed a more discriminative convolutional layer compared to conventional GCN and achieved permutation invariant prediction without losing the capability of capturing complicated interactions.

Despite the fact that deep learning has achieved remarkable success in various domains over the past decade, its application in molecular informatics and drug discovery is still limited. Recent advances in adapting deep architectures to structured data have opened a new paradigm for pharmaceutical research. In this survey, we provide a systematic review on the emerging field of graph convolutional networks and their applications in drug discovery and molecular informatics. Typically we are interested in why and how graph convolution networks can help in drug-related tasks. We elaborate the existing applications through four perspectives: molecular property and activity prediction, interaction prediction, synthesis prediction and de novo drug design. We briefly introduce the theoretical foundations behind graph convolutional networks and illustrate various architectures based on different formulations. Then we summarize the representative applications in drug-related problems. We also discuss the current challenges and future possibilities of applying graph convolutional networks to drug discovery.

MOTIVATION: The use of liquid biopsies for cancer patients enables the non-invasive tracking of treatment response and tumor dynamics through single or serial blood drawn tests. Next-generation sequencing assays allow for the simultaneous interrogation of extended sets of somatic single-nucleotide variants (SNVs) in circulating cell-free DNA (cfDNA), a mixture of DNA molecules originating both from normal and tumor tissue cells. However, low circulating tumor DNA (ctDNA) fractions together with sequencing background noise and potential tumor heterogeneity challenge the ability to confidently call SNVs. RESULTS: We present a computational methodology, called Adaptive Base Error Model in Ultra-deep Sequencing data (ABEMUS), which combines platform-specific genetic knowledge and empirical signal to readily detect and quantify somatic SNVs in cfDNA. We tested the capability of our method to analyze data generated using different platforms with distinct sequencing error properties and we compared ABEMUS performances with other popular SNV callers on both synthetic and real cancer patients sequencing data. Results show that ABEMUS performs better in most of the tested conditions proving its reliability in calling low variant allele frequencies somatic SNVs in low ctDNA levels plasma samples. AVAILABILITY AND IMPLEMENTATION: ABEMUS is cross-platform and can be installed as R package. The source code is maintained on Github at http://github.com/cibiobcg/abemus, and it is also available at CRAN official R repository. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Drug repurposing, identifying novel indications for drugs, bypasses common drug development pitfalls to ultimately deliver therapies to patients faster. However, most repurposing discoveries have been led by anecdotal observations (e.g. Viagra) or experimental-based repurposing screens, which are costly, time-consuming, and imprecise. Recently, more systematic computational approaches have been proposed, however these rely on utilizing the information from the diseases a drug is already approved to treat. This inherently limits the algorithms, making them unusable for investigational molecules. Here, we present a computational approach to drug repurposing, CATNIP, that requires only biological and chemical information of a molecule. CATNIP is trained with 2,576 diverse small molecules and uses 16 different drug similarity features, such as structural, target, or pathway based similarity. This model obtains significant predictive power (AUC = 0.841). Using our model, we created a repurposing network to identify broad scale repurposing opportunities between drug types. By exploiting this network, we identified literature-supported repurposing candidates, such as the use of systemic hormonal preparations for the treatment of respiratory illnesses. Furthermore, we demonstrated that we can use our approach to identify novel uses for defined drug classes. We found that adrenergic uptake inhibitors, specifically amitriptyline and trimipramine, could be potential therapies for Parkinson's disease. Additionally, using CATNIP, we predicted the kinase inhibitor, vandetanib, as a possible treatment for Type 2 Diabetes. Overall, this systematic approach to drug repurposing lays the groundwork to streamline future drug development efforts.

Recent advances in biomedical research have generated a large volume of drug-related data. To effectively handle this flood of data, many initiatives have been taken to help researchers make good use of them. As the results of these initiatives, many drug knowledge bases have been constructed. They range from simple ones with specific focuses to comprehensive ones that contain information on almost every aspect of a drug. These curated drug knowledge bases have made significant contributions to the development of efficient and effective health information technologies for better health-care service delivery. Understanding and comparing existing drug knowledge bases and how they are applied in various biomedical studies will help us recognize the state of the art and design better knowledge bases in the future. In addition, researchers can get insights on novel applications of the drug knowledge bases through a review of successful use cases. In this study, we provide a review of existing popular drug knowledge bases and their applications in drug-related studies. We discuss challenges in constructing and using drug knowledge bases as well as future research directions toward a better ecosystem of drug knowledge bases.

BACKGROUND: To further our understanding of immunopeptidomics, improved tools are needed to identify peptides presented by major histocompatibility complex class I (MHC-I). Many existing tools are limited by their reliance upon chemical affinity data, which is less biologically relevant than sampling by mass spectrometry, and other tools are limited by incomplete exploration of machine learning approaches. Herein, we assemble publicly available data describing human peptides discovered by sampling the MHC-I immunopeptidome with mass spectrometry and use this database to train random forest classifiers (ForestMHC) to predict presentation by MHC-I. RESULTS: As measured by precision in the top 1% of predictions, our method outperforms NetMHC and NetMHCpan on test sets, and it outperforms both these methods and MixMHCpred on new data from an ovarian carcinoma cell line. We also find that random forest scores correlate monotonically, but not linearly, with known chemical binding affinities, and an information-based analysis of classifier features shows the importance of anchor positions for our classification. The random-forest approach also outperforms a deep neural network and a convolutional neural network trained on identical data. Finally, we use our large database to confirm that gene expression partially determines peptide presentation. CONCLUSIONS: ForestMHC is a promising method to identify peptides bound by MHC-I. We have demonstrated the utility of random forest-based approaches in predicting peptide presentation by MHC-I, assembled the largest known database of MS binding data, and mined this database to show the effect of gene expression on peptide presentation. ForestMHC has potential applicability to basic immunology, rational vaccine design, and neoantigen binding prediction for cancer immunotherapy. This method is publicly available for applications and further validation.

SUMMARY: Whole exome sequencing (WES) is widely utilized both in translational cancer genomics studies and in the setting of precision medicine. Stratification of individual's ethnicity is fundamental for the correct interpretation of personal genomic variation impact. We implemented EthSEQ to provide reliable and rapid ethnicity annotation from whole exome sequencing individual's data, validated it on 1000 Genome Project and TCGA data (2700 samples) demonstrating high precision, and finally assessed computational performances compared to other tools. EthSEQ can be integrated into any WES based processing pipeline and exploits multi-core capabilities. AVAILABILITY AND IMPLEMENTATION: R package available at github.com/aromanel/EthSEQ and CRAN repository. CONTACT: [email protected] or [email protected]. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

OBJECTIVE: This paper describes the Precision Medicine Knowledge Base (PMKB; https://pmkb.weill.cornell.edu ), an interactive online application for collaborative editing, maintenance, and sharing of structured clinical-grade cancer mutation interpretations. MATERIALS AND METHODS: PMKB was built using the Ruby on Rails Web application framework. Leveraging existing standards such as the Human Genome Variation Society variant description format, we implemented a data model that links variants to tumor-specific and tissue-specific interpretations. Key features of PMKB include support for all major variant types, standardized authentication, distinct user roles including high-level approvers, and detailed activity history. A REpresentational State Transfer (REST) application-programming interface (API) was implemented to query the PMKB programmatically. RESULTS: At the time of writing, PMKB contains 457 variant descriptions with 281 clinical-grade interpretations. The EGFR, BRAF, KRAS, and KIT genes are associated with the largest numbers of interpretable variants. PMKB's interpretations have been used in over 1500 AmpliSeq tests and 750 whole-exome sequencing tests. The interpretations are accessed either directly via the Web interface or programmatically via the existing API. DISCUSSION: An accurate and up-to-date knowledge base of genomic alterations of clinical significance is critical to the success of precision medicine programs. The open-access, programmatically accessible PMKB represents an important attempt at creating such a resource in the field of oncology. CONCLUSION: The PMKB was designed to help collect and maintain clinical-grade mutation interpretations and facilitate reporting for clinical cancer genomic testing. The PMKB was also designed to enable the creation of clinical cancer genomics automated reporting pipelines via an API.

A promising alternative to address the problem of acquired drug resistance is to rely on combination therapies. Identification of the right combinations is often accomplished through trial and error, a labor and resource intensive process whose scale quickly escalates as more drugs can be combined. To address this problem, we present a broad computational approach for predicting synergistic combinations using easily obtainable single drug efficacy, no detailed mechanistic understanding of drug function, and limited drug combination testing. When applied to mutant BRAF melanoma, we found that our approach exhibited significant predictive power. Additionally, we validated previously untested synergy predictions involving anticancer molecules. As additional large combinatorial screens become available, this methodology could prove to be impactful for identification of drug synergy in context of other types of cancers.

In this review, we describe key components of a computational infrastructure for a precision medicine program that is based on clinical-grade genomic sequencing. Specific aspects covered in this review include software components and hardware infrastructure, reporting, integration into Electronic Health Records for routine clinical use and regulatory aspects. We emphasize informatics components related to reproducibility and reliability in genomic testing, regulatory compliance, traceability and documentation of processes, integration into clinical workflows, privacy requirements, prioritization and interpretation of results to report based on clinical needs, rapidly evolving knowledge base of genomic alterations and clinical treatments and return of results in a timely and predictable fashion. We also seek to differentiate between the use of precision medicine in germline and cancer.

BACKGROUND: Chromatin immunoprecipitation followed by next generation sequencing (ChIP-seq), enables unbiased and genome-wide mapping of protein-DNA interactions and epigenetic marks. The first step in ChIP-seq data analysis involves the identification of peaks (i.e., genomic locations with high density of mapped sequence reads). The next step consists of interpreting the biological meaning of the peaks through their association with known genes, pathways, regulatory elements, and integration with other experiments. Although several programs have been published for the analysis of ChIP-seq data, they often focus on the peak detection step and are usually not well suited for thorough, integrative analysis of the detected peaks. RESULTS: To address the peak interpretation challenge, we have developed ChIPseeqer, an integrative, comprehensive, fast and user-friendly computational framework for in-depth analysis of ChIP-seq datasets. The novelty of our approach is the capability to combine several computational tools in order to create easily customized workflows that can be adapted to the user's needs and objectives. In this paper, we describe the main components of the ChIPseeqer framework, and also demonstrate the utility and diversity of the analyses offered, by analyzing a published ChIP-seq dataset. CONCLUSIONS: ChIPseeqer facilitates ChIP-seq data analysis by offering a flexible and powerful set of computational tools that can be used in combination with one another. The framework is freely available as a user-friendly GUI application, but all programs are also executable from the command line, thus providing flexibility and automatability for advanced users.

BACKGROUND: The accurate determination of transcription factor binding affinities is an important problem in biology and key to understanding the gene regulation process. Position weight matrices are commonly used to represent the binding properties of transcription factor binding sites but suffer from low information content and a large number of false matches in the genome. We describe a novel algorithm for the refinement of position weight matrices representing transcription factor binding sites based on experimental data, including ChIP-chip analyses. We present an iterative weight matrix optimization method that is more accurate in distinguishing true transcription factor binding sites from a negative control set. The initial position weight matrix comes from JASPAR, TRANSFAC or other sources. The main new features are the discriminative nature of the method and matrix width and length optimization. RESULTS: The algorithm was applied to the increasing collection of known transcription factor binding sites obtained from ChIP-chip experiments. The results show that our algorithm significantly improves the sensitivity and specificity of matrix models for identifying transcription factor binding sites. CONCLUSION: When the transcription factor is known, it is more appropriate to use a discriminative approach such as the one presented here to derive its transcription factor-DNA binding properties starting with a matrix, as opposed to performing de novo motif discovery. Generating more accurate position weight matrices will ultimately contribute to a better understanding of eukaryotic transcriptional regulation, and could potentially offer a better alternative to ab initio motif discovery.

We used network-level conservation between pairs of fly (Drosophila melanogaster/D. pseudoobscura) and worm (Caenorhabditis elegans/C. briggsae) genomes to detect highly conserved mRNA motifs in 3' untranslated regions. Many of these elements are complementary to the 5' extremity of known microRNAs (miRNAs), and likely correspond to their target sites. We also identify known targets of RNA-binding proteins, and many novel sites not yet known to be functional. Coherent sets of genes with similar function often bear the same conserved elements, providing new insights into their cellular functions. We also show that target sites for distinct miRNAs are often simultaneously conserved, suggesting combinatorial regulation by multiple miRNAs. A genome-wide search for conserved stem-loops, containing complementary sequences to the novel sites, revealed many new candidate miRNAs that likely target them. We also provide evidence that posttranscriptional networks have undergone extensive rewiring across distant phyla, despite strong conservation of regulatory elements themselves.