Neural Collective Matrix Factorization for integrated analysis of heterogeneous biomedical data.

MOTIVATION: In many biomedical studies, there arises the need to integrate data from multiple directly or indirectly related sources. Collective matrix factorization (CMF) and its variants are models designed to collectively learn from arbitrary collections of matrices. The latent factors learnt are rich integrative representations that can be used in downstream tasks, such as clustering or relation prediction with standard machine-learning models. Previous CMF-based methods have numerous modeling limitations. They do not adequately capture complex non-linear interactions and do not explicitly model varying sparsity and noise levels in the inputs, and some cannot model inputs with multiple datatypes. These inadequacies limit their use on many biomedical datasets. RESULTS: To address these limitations, we develop Neural Collective Matrix Factorization (NCMF), the first fully neural approach to CMF. We evaluate NCMF on relation prediction tasks of gene-disease association prediction and adverse drug event prediction, using multiple datasets. In each case, data are obtained from heterogeneous publicly available databases and used to learn representations to build predictive models. NCMF is found to outperform previous CMF-based methods and several state-of-the-art graph embedding methods for representation learning in our experiments. Our experiments illustrate the versatility and efficacy of NCMF in representation learning for seamless integration of heterogeneous data. AVAILABILITY AND IMPLEMENTATION: https://github.com/ajayago/NCMF_bioinformatics. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

Apical Shear Stress Enhanced Organic Cation Transport in Human OCT2/MATE1-Transfected Madin-Darby Canine Kidney Cells Involves Ciliary Sensing.

Active transport by renal proximal tubules plays a significant role in drug disposition. During drug development, estimates of renal excretion are essential to dose determination. Kidney bioreactors that reproduce physiologic cues in the kidney, such as flow-induced shear stress, may better predict in vivo drug behavior than do current in vitro models. In this study, we investigated the role of shear stress on active transport of 4-(4-(dimethylamino)styryl)-N-methylpyridinium iodide (ASP+) by Madin-Darby canine kidney cells exogenously expressing the human organic cation transporters organic cation transporter 2 (OCT2) and multidrug and toxin extrusion protein 1 (MATE1). Cells cultured in a parallel plate under continuous media perfusion formed a tight monolayer with a high barrier to inulin. In response to increasing levels of shear stress (0.2-2 dynes/cm2), cells showed a corresponding increase in transport of ASP+, reaching a maximal 4.2-fold increase at 2 dynes/cm2 compared with cells cultured under static conditions. This transport was inhibited with imipramine, indicating active transport was present under shear stress conditions. Cells exposed to shear stress of 2 dynes/cm2 also showed an increase in RNA expression of both transfected human and endogenous OCT2 (3.7- and 2.0-fold, respectively). Removal of cilia by ammonium sulfate eliminated the effects of shear on ASP+ transport at 0.5 dynes/cm2 with no effect on ASP+ transport under static conditions. These results indicate that shear stress affects active transport of organic cations in renal tubular epithelial cells in a cilia-dependent manner.

ASTER: A Method to Predict Clinically Relevant Synthetic Lethal Genetic Interactions.

A Synthetic Lethal (SL) interaction is a functional relationship between two genes or functional entities where the loss of either entity is viable but the loss of both is lethal. Such pairs can be used to develop targeted anticancer therapies with fewer side effects and reduced overtreatment. However, finding clinically relevant SL interactions remains challenging. Leveraging unified gene expression data of both disease-free and cancerous samples, we design a new technique based on statistical hypothesis testing, called ASTER, to identify SL pairs. We empirically find that the patterns of mutually exclusivity ASTER finds using genomic and transcriptomic data provides a strong signal of synthetic lethality. For large-scale multiple hypothesis testing, we develop an extension called ASTER++ that can utilize additional input gene features within the hypothesis testing framework. Our computational and functional experiments demonstrate the efficacy of ASTER in identifying SL pairs with potential therapeutic benefits.

Molecular profiling of metastatic breast cancer and target-based therapeutic matching in an Asian tertiary phase I oncology unit.

Introduction: Molecular profiling of metastatic breast cancer (MBC) through the widespread use of next-generation sequencing (NGS) has highlighted actionable mutations and driven trials of targeted therapy matched to tumour molecular profiles, with improved outcomes reported using such an approach. Here, we review NGS results and treatment outcomes for a cohort of Asian MBC patients in the phase I unit of a tertiary centre. Methods: Patients with MBC referred to a phase I unit underwent NGS via Ion AmpliSeq Cancer Hotspot v2 (ACH v2, 2014-2017) prior to institutional change to FoundationOne CDx (FM1; 2017-2022). Patients were counselled on findings and enrolled on matched therapeutic trials, where available. Outcomes for all subsequent treatment events were recorded to data cut-off on January 31, 2022. Results: A total of 215 patients were enrolled with successful NGS in 158 patients. The PI3K/AKT/PTEN pathway was the most altered with one or more of the pathway member genes PIK3/AKT/PTEN affected in 62% (98/158) patients and 43% of tumours harbouring a PIK3CA alteration. Tumour mutational burden (TMB) was reported in 96/109 FM1 sequenced patients, with a mean TMB of 5.04 mt/Mb and 13% (12/96) with TMB ≥ 10 mt/Mb. Treatment outcomes were evaluable in 105/158 patients, with a pooled total of 216 treatment events recorded. Matched treatment was administered in 47/216 (22%) events and associated with prolonged median progression-free survival (PFS) of 21.0 weeks [95% confidence interval (CI) 11.7, 26.0 weeks] versus 12.1 weeks (95% CI 10.0, 15.4 weeks) in unmatched, with hazard ratio (HR) for progression or death of 0.63 (95% CI 0.41, 0.97; p = 0.034). In the subgroup of PIK3/AKT/PTEN-altered MBC, the HR for progression or death was 0.57 (95% CI 0.35, 0.92; p = 0.02), favouring matched treatment. Per-patient overall survival (OS) analysis (n = 105) showed improved survival for patients receiving matched treatment versus unmatched, with median OS (mOS) of 30.1 versus 11.8 months, HR = 0.45 (95% CI 0.24, 0.84; p = 0.013). Objective response rate (ORR) in the overall population was similar in matched and unmatched treatment events (23.7% versus 17.2%, odds ratio of response 1.14 95% CI 0.50, 2.62; p = 0.75). Conclusions: Broad-panel NGS in MBC is feasible, allowing therapeutic matching, which was associated with improvements in PFS and OS.

Nanoengineered surfaces enhance drug loading and adhesion.

To circumvent the barriers encountered by macromolecules at the gastrointestinal mucosa, sufficient therapeutic macromolecules must be delivered in close proximity to cells.(1) Previously, we have shown that silicon nanowires penetrate the mucous layer and adhere directly to cells under high shear.(2) In this work, we characterize potential reservoirs and load macromolecules into interstitial space between nanowires. We show significant increases in loading capacity due to nanowires while retaining adhesion of loaded particles under high shear.

Adverse Drug Event Prediction Using Noisy Literature-Derived Knowledge Graphs: Algorithm Development and Validation.

BACKGROUND: Adverse drug events (ADEs) are unintended side effects of drugs that cause substantial clinical and economic burdens globally. Not all ADEs are discovered during clinical trials; therefore, postmarketing surveillance, called pharmacovigilance, is routinely conducted to find unknown ADEs. A wealth of information, which facilitates ADE discovery, lies in the growing body of biomedical literature. Knowledge graphs (KGs) encode information from the literature, where the vertices and the edges represent clinical concepts and their relations, respectively. The scale and unstructured form of the literature necessitates the use of natural language processing (NLP) to automatically create such KGs. Previous studies have demonstrated the utility of such literature-derived KGs in ADE prediction. Through unsupervised learning of the representations (features) of clinical concepts from the KG, which are used in machine learning models, state-of-the-art results for ADE prediction were obtained on benchmark data sets. OBJECTIVE: Due to the use of NLP to infer literature-derived KGs, there is noise in the form of false positive (erroneous) and false negative (absent) nodes and edges. Previous representation learning methods do not account for such inaccuracies in the graph. NLP algorithms can quantify the confidence in their inference of extracted concepts and relations from the literature. Our hypothesis, which motivates this work, is that by using such confidence scores during representation learning, the learned embeddings would yield better features for ADE prediction models. METHODS: We developed methods to use these confidence scores on two well-known representation learning methods-DeepWalk and Translating Embeddings for Modeling Multi-relational Data (TransE)-to develop their weighted versions: Weighted DeepWalk and Weighted TransE. These methods were used to learn representations from a large literature-derived KG, the Semantic MEDLINE Database, which contains more than 93 million clinical relations. They were compared with Embedding of Semantic Predications, which, to our knowledge, is the best reported representation learning method using the Semantic MEDLINE Database with state-of-the-art results for ADE prediction. Representations learned from different methods were used (separately) as features of drugs and diseases to build classification models for ADE prediction using benchmark data sets. The methods were compared rigorously over multiple cross-validation settings. RESULTS: The weighted versions we designed were able to learn representations that yielded more accurate predictive models than the corresponding unweighted versions of both DeepWalk and TransE, as well as Embedding of Semantic Predications, in our experiments. There were performance improvements of up to 5.75% in the F1-score and 8.4% in the area under the receiver operating characteristic curve value, thus advancing the state of the art in ADE prediction from literature-derived KGs. CONCLUSIONS: Our classification models can be used to aid pharmacovigilance teams in detecting potentially new ADEs. Our experiments demonstrate the importance of modeling inaccuracies in the inferred KGs for representation learning.