Supervised prediction of drug-induced nephrotoxicity based on interleukin-6 and -8 expression levels

In silico prediction of drug-induced nephrotoxicity: current progress and pitfalls

INTRODUCTION

Due to its role in absorption and metabolism, the kidney is an important target for drug toxicity. Drug-induced nephrotoxicity (DIN) presents a significant challenge in clinical practice and drug development. Conventional methods for assessing nephrotoxicity have limitations, highlighting the need for innovative approaches. In recent years, in silico methods have emerged as promising tools for predicting DIN.

AREAS COVERED

A literature search was performed using PubMed and Web of Science, from 2013 to February 2023 for this review. This review provides an overview of the current progress and pitfalls in the in silico prediction of DIN, which discusses the principles and methodologies of computational models.

EXPERT OPINION

Despite significant advancements, this review identified issues accentuates the pivotal imperatives of data fidelity, model optimization, interdisciplinary collaboration, and mechanistic comprehension in sculpting the vista of DIN prediction. Integration of multiple data sources and collaboration between disciplines are essential for improving predictive models. Ultimately, a holistic approach combining computational, experimental, and clinical methods will enhance our understanding and management of DIN.

A proof-of-concept assay for quantitative and optical assessment of drug-induced toxicity in renal organoids

Kidneys are complex organs, and reproducing their function and physiology in a laboratory setting remains difficult. During drug development, potential compounds may exhibit unexpected nephrotoxic effects, which imposes a significant financial burden on pharmaceutical companies. As a result, there is an ongoing need for more accurate model systems. The use of renal organoids to simulate responses to nephrotoxic insults has the potential to bridge the gap between preclinical drug efficacy studies in cell cultures and animal models, and the stages of clinical trials in humans. Here we established an accessible fluorescent whole-mount approach for nuclear and membrane staining to first provide an overview of the organoid histology. Furthermore, we investigated the potential of renal organoids to model responses to drug toxicity. For this purpose, organoids were treated with the chemotherapeutic agent doxorubicin for 48 h. When cell viability was assessed biochemically, the organoids demonstrated a significant, dose-dependent decline in response to the treatment. Confocal microscopy revealed visible tubular disintegration and a loss of cellular boundaries at high drug concentrations. This observation was further reinforced by a dose-dependent decrease of the nuclear area in the analyzed images. In contrast to other approaches, in this study, we provide a straightforward experimental framework for drug toxicity assessment in renal organoids that may be used in early research stages to assist screen for potential adverse effects of compounds.

The Synergy between Deep Learning and Organs-on-Chips for High-Throughput Drug Screening: A Review

Organs-on-chips (OoCs) are miniature microfluidic systems that have arguably become a class of advanced in vitro models. Deep learning, as an emerging topic in machine learning, has the ability to extract a hidden statistical relationship from the input data. Recently, these two areas have become integrated to achieve synergy for accelerating drug screening. This review provides a brief description of the basic concepts of deep learning used in OoCs and exemplifies the successful use cases for different types of OoCs. These microfluidic chips are of potential to be assembled as highly potent human-on-chips with complex physiological or pathological functions. Finally, we discuss the future supply with perspectives and potential challenges in terms of combining OoCs and deep learning for image processing and automation designs.

Kidney organoids: current knowledge and future directions

The kidney is a highly complex organ in the human body. Although creating an in vitro model of the human kidney is challenging, tremendous advances have been made in recent years. Kidney organoids are in vitro kidney models that are generated from stem cells in three-dimensional (3D) cultures. They exhibit remarkable degree of similarities with the native tissue in terms of cell type, morphology, and function. The establishment of 3D kidney organoids facilitates a mechanistic study of cell communications, and these organoids can be used for drug screening, disease modeling, and regenerative medicine applications. This review discusses the cellular complexity during in vitro kidney generation. We intend to highlight recent progress in kidney organoids and the applications of these relatively new technologies.

Prediction of drug-induced kidney injury in drug discovery

Drug induced kidney injury is one of the leading causes of failure of drug development programs in the clinic. Early prediction of renal toxicity potential of drugs is crucial to the success of drug candidates in the clinic. The dynamic nature of the functioning of the kidney and the presence of drug uptake proteins introduce additional challenges in the prediction of renal injury caused by drugs. Renal injury due to drugs can be caused by a wide variety of mechanisms and can be broadly classified as toxic or obstructive. Several biomarkers are available for in vitro and in vivo detection of renal injury. In vitro static and dynamic (microfluidic) cellular models and preclinical models can provide valuable information regarding the toxicity potential of drugs. Differences in pharmacology and subsequent disconnect in biomarker response, differences in the expression of transporter and enzyme proteins between in vitro to in vivo systems and between preclinical species and humans are some of the limitations of current experimental models. The progress in microfluidic (kidney-on-chip) platforms in combination with the ability of 3-dimensional cell culture can help in addressing some of these issues in the future. Finally, newer in silico and computational techniques like physiologically based pharmacokinetic modeling and machine learning have demonstrated potential in assisting prediction of drug induced kidney injury.

Nephrotoxic antisense oligonucleotide SPC5001 induces kidney injury biomarkers in a proximal tubule-on-a-chip

Antisense oligonucleotides (ASOs) are a promising therapeutic modality. However, failure to predict acute kidney injury induced by SPC5001 ASO observed in a clinical trial suggests the need for additional preclinical models to complement the preceding animal toxicity studies. To explore the utility of in vitro systems in this space, we evaluated the induction of nephrotoxicity and kidney injury biomarkers by SPC5001 in human renal proximal tubule epithelial cells (HRPTEC), cultured in 2D, and in a recently developed kidney proximal tubule-on-a-chip. 2D HRPTEC cultures were exposed to the nephrotoxic ASO SPC5001 or the safe control ASO 556089 (0.16-40 µM) for up to 72 h, targeting PCSK9 and MALAT1, respectively. Both ASOs induced a concentration-dependent downregulation of their respective mRNA targets but cytotoxicity (determined by LDH activity) was not observed at any concentration. Next, chip-cultured HRPTEC were exposed to SPC5001 (0.5 and 5 µM) and 556089 (1 and 10 µM) for 48 h to confirm downregulation of their respective target transcripts, with 74.1 ± 5.2% for SPC5001 (5 µM) and 79.4 ± 0.8% for 556089 (10 µM). During extended exposure for up to 20 consecutive days, only SPC5001 induced cytotoxicity (at the higher concentration; 5 µM), as evaluated by LDH in the perfusate medium. Moreover, perfusate levels of biomarkers KIM-1, NGAL, clusterin, osteopontin and VEGF increased 2.5 ± 0.2-fold, 3.9 ± 0.9-fold, 2.3 ± 0.6-fold, 3.9 ± 1.7-fold and 1.9 ± 0.4-fold respectively, in response to SPC5001, generating distinct time-dependent profiles. In conclusion, target downregulation, cytotoxicity and kidney injury biomarkers were induced by the clinically nephrotoxic ASO SPC5001, demonstrating the translational potential of this kidney on-a-chip.

The Matthews correlation coefficient (MCC) is more reliable than balanced accuracy, bookmaker informedness, and markedness in two-class confusion matrix evaluation

Evaluating binary classifications is a pivotal task in statistics and machine learning, because it can influence decisions in multiple areas, including for example prognosis or therapies of patients in critical conditions. The scientific community has not agreed on a general-purpose statistical indicator for evaluating two-class confusion matrices (having true positives, true negatives, false positives, and false negatives) yet, even if advantages of the Matthews correlation coefficient (MCC) over accuracy and F1 score have already been shown.In this manuscript, we reaffirm that MCC is a robust metric that summarizes the classifier performance in a single value, if positive and negative cases are of equal importance. We compare MCC to other metrics which value positive and negative cases equally: balanced accuracy (BA), bookmaker informedness (BM), and markedness (MK). We explain the mathematical relationships between MCC and these indicators, then show some use cases and a bioinformatics scenario where these metrics disagree and where MCC generates a more informative response.Additionally, we describe three exceptions where BM can be more appropriate: analyzing classifications where dataset prevalence is unrepresentative, comparing classifiers on different datasets, and assessing the random guessing level of a classifier. Except in these cases, we believe that MCC is the most informative among the single metrics discussed, and suggest it as standard measure for scientists of all fields. A Matthews correlation coefficient close to +1, in fact, means having high values for all the other confusion matrix metrics. The same cannot be said for balanced accuracy, markedness, bookmaker informedness, accuracy and F1 score.

Machine Learning in Predictive Toxicology: Recent Applications and Future Directions for Classification Models

In recent times, machine learning has become increasingly prominent in predictive toxicology as it has shifted from in vivo studies toward in silico studies. Currently, in vitro methods together with other computational methods such as quantitative structure-activity relationship modeling and absorption, distribution, metabolism, and excretion calculations are being used. An overview of machine learning and its applications in predictive toxicology is presented here, including support vector machines (SVMs), random forest (RF) and decision trees (DTs), neural networks, regression models, naïve Bayes, k-nearest neighbors, and ensemble learning. The recent successes of these machine learning methods in predictive toxicology are summarized, and a comparison of some models used in predictive toxicology is presented. In predictive toxicology, SVMs, RF, and DTs are the dominant machine learning methods due to the characteristics of the data available. Lastly, this review describes the current challenges facing the use of machine learning in predictive toxicology and offers insights into the possible areas of improvement in the field.

3D kidney organoids for bench-to-bedside translation

The kidneys are essential organs that filter the blood, removing urinary waste while maintaining fluid and electrolyte homeostasis. Current conventional research models such as static cell cultures and animal models are insufficient to grasp the complex human in vivo situation or lack translational value. To accelerate kidney research, novel research tools are required. Recent developments have allowed the directed differentiation of induced pluripotent stem cells to generate kidney organoids. Kidney organoids resemble the human kidney in vitro and can be applied in regenerative medicine and as developmental, toxicity, and disease models. Although current studies have shown great promise, challenges remain including the immaturity, limited reproducibility, and lack of perfusable vascular and collecting duct systems. This review gives an overview of our current understanding of nephrogenesis that enabled the generation of kidney organoids. Next, the potential applications of kidney organoids are discussed followed by future perspectives. This review proposes that advancement in kidney organoid research will be facilitated through our increasing knowledge on nephrogenesis and combining promising techniques such as organ-on-a-chip models.

Artificial intelligence in drug combination therapy

Currently, the development of medicines for complex diseases requires the development of combination drug therapies. It is necessary because in many cases, one drug cannot target all necessary points of intervention. For example, in cancer therapy, a physician often meets a patient having a genomic profile including more than five molecular aberrations. Drug combination therapy has been an area of interest for a while, for example the classical work of Loewe devoted to the synergism of drugs was published in 1928-and it is still used in calculations for optimal drug combinations. More recently, over the past several years, there has been an explosion in the available information related to the properties of drugs and the biomedical parameters of patients. For the drugs, hundreds of 2D and 3D molecular descriptors for medicines are now available, while for patients, large data sets related to genetic/proteomic and metabolomics profiles of the patients are now available, as well as the more traditional data relating to the histology, history of treatments, pretreatment state of the organism, etc. Moreover, during disease progression, the genetic profile can change. Thus, the ability to optimize drug combinations for each patient is rapidly moving beyond the comprehension and capabilities of an individual physician. This is the reason, that biomedical informatics methods have been developed and one of the more promising directions in this field is the application of artificial intelligence (AI). In this review, we discuss several AI methods that have been successfully implemented in several instances of combination drug therapy from HIV, hypertension, infectious diseases to cancer. The data clearly show that the combination of rule-based expert systems with machine learning algorithms may be promising direction in this field.

Developing a Multi-Dose Computational Model for Drug-Induced Hepatotoxicity Prediction Based on Toxicogenomics Data

Drug-induced hepatotoxicity may cause acute and chronic liver disease, leading to great concern for patient safety. It is also one of the main reasons for drug withdrawal from the market. Toxicogenomics data has been widely used in hepatotoxicity prediction. In our study, we proposed a multi-dose computational model to predict the drug-induced hepatotoxicity based on gene expression and toxicity data. The dose/concentration information after drug treatment is fully utilized in our study based on the dose-response curve, thus a more informative representative of the dose-response relationship is considered. We also proposed a new feature selection method, named MEMO, which is also one important aspect of our multi-dose model in our study, to deal with the high-dimensional toxicogenomics data. We validated the proposed model using the TG-GATEs, which is a large database recording toxicogenomics data from multiple views. The experimental results show that the drug-induced hepatotoxicity can be predicted with high accuracy and efficiency using the proposed predictive model.

Understanding the effect of repeated administration of meloxicam on feline renal cortex and medulla: A lipidomics and metabolomics approach

Repeated administration of meloxicam can cause kidney damage in cats by mechanisms that remain unclear. Metabolomics and lipidomics are powerful, noninvasive approaches used to investigate tissue response to drug exposure. Thus, the objective of this study was to assess the effects of meloxicam on the feline kidney using untargeted metabolomics and lipidomics approaches. Female young-adult purpose-breed cats were allocated into the control (n = 4) and meloxicam (n = 4) groups. Cats in the control and meloxicam groups were treated daily with saline and meloxicam at 0.3 mg/kg subcutaneously for 17 days, respectively. Renal cortices and medullas were collected at the end of the treatment period. Random forest and metabolic pathway analyses were used to identify metabolites that discriminate meloxicam-treated from saline-treated cats and to identify disturbed metabolic pathways in renal tissue. Our results revealed that the repeated administration of meloxicam to cats altered the kidney metabolome and lipidome and suggest that at least 40 metabolic pathways were altered in the renal cortex and medulla. These metabolic pathways included lipid, amino acid, carbohydrate, nucleotide and energy metabolisms, and metabolism of cofactors and vitamins. This is the first study using a pharmacometabonomics approach for studying the molecular effects of meloxicam on feline kidneys.

Road Map for Development of Stem Cell-Based Alternative Test Methods

Much progress has been made in establishing strategies for differentiation of induced human pluripotent stem cells (hiPSCs). However, differentiated hiPSCs are not yet routinely used for prediction of toxicity. Here, limiting factors are summarised and possibilities for improvement are discussed, with a focus on hepatocytes, cardiomyocytes, tubular epithelial cells, and developmental toxicity. Moreover, we make recommendations for further fine-tuning of differentiation protocols for hiPSCs to hepatocyte-like cells by comparing individual steps of currently available protocols to the mechanisms occurring during embryonic development. A road map is proposed to facilitate test system development, including a description of the most useful performance metrics.

Kidney organoids in translational medicine: Disease modeling and regenerative medicine

The kidney is one of the most complex organs composed of multiple cell types, functioning to maintain homeostasis by means of the filtering of metabolic wastes, balancing of blood electrolytes, and adjustment of blood pressure. Recent advances in 3D culture technologies in vitro enabled the generation of "organoids" which mimic the structure and function of in vivo organs. Organoid technology has allowed for new insights into human organ development and human pathophysiology, with great potential for translational research. Increasing evidence shows that kidney organoids are a useful platform for disease modeling of genetic kidney diseases when derived from genetic patient iPSCs and/or CRISPR-mutated stem cells. Although single cell RNA-seq studies highlight the technical difficulties underlying kidney organoid generation reproducibility and variation in differentiation protocols, kidney organoids still hold great potential to understand kidney pathophysiology as applied to kidney injury and fibrosis. In this review, we summarize various studies of kidney organoids, disease modeling, genome-editing, and bioengineering, and additionally discuss the potential of and current challenges to kidney organoid research.

Repeated administration of the NSAID meloxicam alters the plasma and urine lipidome

Non-steroidal anti-inflammatories (NSAIDs), such as meloxicam, are the mainstay for treating painful and inflammatory conditions in animals and humans; however, the repeated administration of NSAIDs can cause adverse effects, limiting the long-term administration of these drugs to some patients. The primary aim of this study was to determine the effects of repeated meloxicam administration on the feline plasma and urine lipidome. Cats (n = 12) were treated subcutaneously with either saline solution or 0.3 mg/kg body weight of meloxicam daily for up to 31 days. Plasma and urine lipidome were determined by LC-MS before the first treatment and at 4, 9 and 13 and 17 days after the first administration of meloxicam. The repeated administration of meloxicam altered the feline plasma and urine lipidome as demonstrated by multivariate statistical analysis. The intensities of 94 out of 195 plasma lipids were altered by the repeated administration of meloxicam to cats (p < 0.05). Furthermore, we identified 12 lipids in plasma and 10 lipids in urine that could serve as biomarker candidates for discriminating animals receiving NSAIDs from healthy controls. Expanding our understanding about the effects of NSAIDs in the body could lead to the discovery of mechanism(s) associated with intolerance to NSAIDs.

Comment on Sjögren et al. (2018) A novel multi-parametric high-content screening assay in ciPTEC-OAT1 to predict drug-induced nephrotoxicity in drug discovery. Arch Toxicol 92(10):3175-3190

Rapid progress is made in the development of high-content screening assays for the prediction of nephrotoxicity. The findings from different laboratories are consistent with respect to endpoints and concentration ranges screened. Discrepancies regarding compound annotation and the predictive performance analysis are discussed.

Applications of kidney organoids derived from human pluripotent stem cells

The establishment of protocols to differentiate kidney organoids from human pluripotent stem cells provides potential applications of kidney organoids in regenerative medicine. Modeling of renal diseases, drug screening, nephrotoxicity testing of compounds, and regenerative therapy are attractive applications. Although much progress still remains to be made in the development of kidney organoids, recent advances in clustered regularly interspaced short palindromic repeat (CRISPR)-CRISPR-associated system 9 (Cas9) genome editing and three-dimensional bioprinting technologies have contributed to the application of kidney organoids in clinical fields. In this section, we review recent advances in the applications of kidney organoids to kidney disease modelling, drug screening, nephrotoxicity testing, and regenerative therapy.

High-throughput prediction of nephrotoxicity in humans

The Lush Science Prize 2016 was awarded to Daniele Zink and Lit-Hsin Loo for the interdisciplinary and collaborative work between their research groups in developing alternative methods for the prediction of nephrotoxicity in humans. The collaboration has led to the establishment of a series of pioneering alternative methods for nephrotoxicity prediction, which includes: predictive gene expression markers based on pro-inflammatory responses; predictive in vitro cellular models based on pluripotent stem cell-derived proximal tubular-like cells; and predictive cellular phenotypic markers based on chromatin and cytoskeletal changes. A high-throughput method was established for chemical testing, which is currently being used to predict the potential human nephrotoxicity of ToxCast compounds in collaboration with the US Environmental Protection Agency. Similar high-throughput imaging-based methodologies are currently being developed and adapted by the Zink and Loo groups, to include other human organs and cell types. The ultimate goal is to develop a portfolio of methods accepted for the accurate prediction of human organ-specific toxicity and the consequent replacement of animal experiments.

Organ/body-on-a-chip based on microfluidic technology for drug discovery

Although animal experiments are indispensable for preclinical screening in the drug discovery process, various issues such as ethical considerations and species differences remain. To solve these issues, cell-based assays using human-derived cells have been actively pursued. However, it remains difficult to accurately predict drug efficacy, toxicity, and organs interactions, because cultivated cells often do not retain their original organ functions and morphologies in conventional in vitro cell culture systems. In the μTAS research field, which is a part of biochemical engineering, the technologies of organ-on-a-chip, based on microfluidic devices built using microfabrication, have been widely studied recently as a novel in vitro organ model. Since it is possible to physically and chemically mimic the in vitro environment by using microfluidic device technology, maintenance of cellular function and morphology, and replication of organ interactions can be realized using organ-on-a-chip devices. So far, functions of various organs and tissues, such as the lung, liver, kidney, and gut have been reproduced as in vitro models. Furthermore, a body-on-a-chip, integrating multi organ functions on a microfluidic device, has also been proposed for prediction of organ interactions. We herein provide a background of microfluidic systems, organ-on-a-chip, Body-on-a-chip technologies, and their challenges in the future.

Regenerative Medicine, Disease Modeling, and Drug Discovery in Human Pluripotent Stem Cell-derived Kidney Tissue

The multitude of research clarifying critical factors in embryonic organ development has been instrumental in human stem cell research. Mammalian organogenesis serves as the archetype for directed differentiation protocols, subdividing the process into a series of distinct intermediate stages that can be chemically induced and monitored for the expression of stage-specific markers. Significant advances over the past few years include established directed differentiation protocols of human embryonic stem cells (hESCs) and human induced pluripotent stem cells (hiPSCs) into human kidney organoids in vitro. Human kidney tissue in vitro simulate the in vivo response when subject to nephrotoxins, providing a novel screening platform during drug discovery to facilitate identification of lead candidates, reduce developmental expenditures, and reduce future rates of drug-induced acute kidney injury. Patient-derived hiPSCs, which bear naturally occurring DNA mutations, may allow for modeling of human genetic diseases to determine pathologic mechanisms and screen for novel therapeutics. In addition, recent advances in genome editing with CRISPR/Cas9 enable to generate specific mutations to study genetic disease with non-mutated lines serving as an ideal isogenic control. The growing population of patients with end-stage kidney disease (ESKD) is a world-wide healthcare problem with higher morbidity and mortality that warrants the discovery of novel forms of renal replacement therapy. Coupling the outlined advances in hiPSC research with innovative bioengineering techniques, such as decellularized kidney and 3D printed scaffolds, may contribute to the development of bioengineered transplantable human kidney tissue as a means of renal replacement therapy.

Stem cell-derived kidney cells and organoids: Recent breakthroughs and emerging applications

The global rise in the numbers of kidney patients and the shortage in transplantable organs have led to an increasing interest in kidney-specific regenerative therapies, renal disease modelling and bioartificial kidneys. Sources for large quantities of high-quality renal cells and tissues would be required, also for applications in in vitro platforms for compound safety and efficacy screening. Stem cell-based approaches for the generation of renal-like cells and tissues would be most attractive, but such methods were not available until recently. This situation has drastically changed since 2013, and various protocols for the generation of renal-like cells and precursors from pluripotent stem cells (PSC) have been established. The most recent breakthroughs were related to the establishment of various protocols for the generation of PSC-derived kidney organoids. In combination with recent advances in genome editing, bioprinting and the establishment of predictive renal screening platforms this results in exciting new possibilities. This review will give a comprehensive overview over current PSC-based protocols for the generation of renal-like cells, precursors and organoids, and their current and potential applications in regenerative medicine, compound screening, disease modelling and bioartificial organs.

A Role for 3D Printing in Kidney-on-a-Chip Platforms

The advancement of "kidney-on-a-chip" platforms - submillimeter-scale fluidic systems designed to recapitulate renal functions in vitro - directly impacts a wide range of biomedical fields, including drug screening, cell and tissue engineering, toxicity testing, and disease modelling. To fabricate kidney-on-a-chip technologies, researchers have primarily adapted traditional micromachining techniques that are rooted in the integrated circuit industry; hence the term, "chip." A significant challenge, however, is that such methods are inherently monolithic, which limits one's ability to accurately recreate the geometric and architectural complexity of the kidney in vivo. Better reproduction of the anatomical complexity of the kidney will allow for more instructive modelling of physiological and pathophysiological events. Emerging additive manufacturing or "three-dimensional (3D) printing" techniques could provide a promising alternative to conventional methodologies. In this article, we discuss recent progress in the development of both kidney-on-a-chip platforms and state-of-the-art submillimeter-scale 3D printing methods, with a focus on biophysical and architectural capabilities. Lastly, we examine the potential for 3D printing-based approaches to extend the efficacy of kidney-on-a-chip systems.

High-throughput imaging-based nephrotoxicity prediction for xenobiotics with diverse chemical structures

The kidney is a major target for xenobiotics, which include drugs, industrial chemicals, environmental toxicants and other compounds. Accurate methods for screening large numbers of potentially nephrotoxic xenobiotics with diverse chemical structures are currently not available. Here, we describe an approach for nephrotoxicity prediction that combines high-throughput imaging of cultured human renal proximal tubular cells (PTCs), quantitative phenotypic profiling, and machine learning methods. We automatically quantified 129 image-based phenotypic features, and identified chromatin and cytoskeletal features that can predict the human in vivo PTC toxicity of 44 reference compounds with ~82 % (primary PTCs) or 89 % (immortalized PTCs) test balanced accuracies. Surprisingly, our results also revealed that a DNA damage response is commonly induced by different PTC toxicants that have diverse chemical structures and injury mechanisms. Together, our results show that human nephrotoxicity can be predicted with high efficiency and accuracy by combining cell-based and computational methods that are suitable for automation.

Prediction of drug-induced nephrotoxicity and injury mechanisms with human induced pluripotent stem cell-derived cells and machine learning methods

The renal proximal tubule is a main target for drug-induced toxicity. The prediction of proximal tubular toxicity during drug development remains difficult. Any in vitro methods based on induced pluripotent stem cell-derived renal cells had not been developed, so far. Here, we developed a rapid 1-step protocol for the differentiation of human induced pluripotent stem cells (hiPSC) into proximal tubular-like cells. These proximal tubular-like cells had a purity of >90% after 8 days of differentiation and could be directly applied for compound screening. The nephrotoxicity prediction performance of the cells was determined by evaluating their responses to 30 compounds. The results were automatically determined using a machine learning algorithm called random forest. In this way, proximal tubular toxicity in humans could be predicted with 99.8% training accuracy and 87.0% test accuracy. Further, we studied the underlying mechanisms of injury and drug-induced cellular pathways in these hiPSC-derived renal cells, and the results were in agreement with human and animal data. Our methods will enable the development of personalized or disease-specific hiPSC-based renal in vitro models for compound screening and nephrotoxicity prediction.

InCoB2014: bioinformatics to tackle the data to knowledge challenge. Introduction

Since 2006, the International Conference on Bioinformatics (InCoB) has been publishing selected papers in BMC Bioinformatics. Papers within the scope of the journal from the 13th InCoB July 31-2 August, 2014 in Sydney, Australia have been compiled in this supplement. These span protein and proteome informatics, structural bioinformatics, software development and bioimaging to pharmacoinformatics and disease informatics, representing the breadth of bioinformatics research in the Asia-Pacific.