In silico Pathway Activation Network Decomposition Analysis (iPANDA) as a method for biomarker development

PandaOmics: An AI-Driven Platform for Therapeutic Target and Biomarker Discovery

PandaOmics is a cloud-based software platform that applies artificial intelligence and bioinformatics techniques to multimodal omics and biomedical text data for therapeutic target and biomarker discovery. PandaOmics generates novel and repurposed therapeutic target and biomarker hypotheses with the desired properties and is available through licensing or collaboration. Targets and biomarkers generated by the platform were previously validated in both in vitro and in vivo studies. PandaOmics is a core component of Insilico Medicine's Pharma.ai drug discovery suite, which also includes Chemistry42 for the de novo generation of novel small molecules, and inClinico─a data-driven multimodal platform that forecasts a clinical trial's probability of successful transition from phase 2 to phase 3. In this paper, we demonstrate how the PandaOmics platform can efficiently identify novel molecular targets and biomarkers for various diseases.

Acquired resistance to immunotherapy and chemoradiation in MYC amplified head and neck cancer

The proto-oncogene MYC encodes a nuclear transcription factor that has an important role in a variety of cellular processes, such as cell cycle progression, proliferation, metabolism, adhesion, apoptosis, and therapeutic resistance. MYC amplification is consistently observed in aggressive forms of several solid malignancies and correlates with poor prognosis and distant metastases. While the tumorigenic effects of MYC in patients with head and neck squamous cell carcinoma (HNSCC) are well known, the molecular mechanisms by which the amplification of this gene may confer treatment resistance, especially to immune checkpoint inhibitors, remains under-investigated. Here we present a unique case of a patient with recurrent/metastatic (R/M) HNSCC who, despite initial response to nivolumab-based treatment, developed rapidly progressive metastatic disease after the acquisition of MYC amplification. We conducted comparative transcriptomic analysis of this patient's tumor at baseline and upon progression to interrogate potential molecular processes through which MYC may confer resistance to immunotherapy and/or chemoradiation and used TCGA-HNSC dataset and an institutional cohort to further explore clinicopathologic features and key molecular networks associated with MYC amplification in HNSCC. This study highlights MYC amplification as a potential mechanism of immune checkpoint inhibitor resistance and suggest its use as a predictive biomarker and potential therapeutic target in R/M HNSCC.

The Role of AI in Drug Discovery

The emergence of Artificial Intelligence (AI) in drug discovery marks a pivotal shift in pharmaceutical research, blending sophisticated computational techniques with conventional scientific exploration to break through enduring obstacles. This review paper elucidates the multifaceted applications of AI across various stages of drug development, highlighting significant advancements and methodologies. It delves into AI's instrumental role in drug design, polypharmacology, chemical synthesis, drug repurposing, and the prediction of drug properties such as toxicity, bioactivity, and physicochemical characteristics. Despite AI's promising advancements, the paper also addresses the challenges and limitations encountered in the field, including data quality, generalizability, computational demands, and ethical considerations. By offering a comprehensive overview of AI's role in drug discovery, this paper underscores the technology's potential to significantly enhance drug development, while also acknowledging the hurdles that must be overcome to fully realize its benefits.

A small-molecule TNIK inhibitor targets fibrosis in preclinical and clinical models

Idiopathic pulmonary fibrosis (IPF) is an aggressive interstitial lung disease with a high mortality rate. Putative drug targets in IPF have failed to translate into effective therapies at the clinical level. We identify TRAF2- and NCK-interacting kinase (TNIK) as an anti-fibrotic target using a predictive artificial intelligence (AI) approach. Using AI-driven methodology, we generated INS018_055, a small-molecule TNIK inhibitor, which exhibits desirable drug-like properties and anti-fibrotic activity across different organs in vivo through oral, inhaled or topical administration. INS018_055 possesses anti-inflammatory effects in addition to its anti-fibrotic profile, validated in multiple in vivo studies. Its safety and tolerability as well as pharmacokinetics were validated in a randomized, double-blinded, placebo-controlled phase I clinical trial (NCT05154240) involving 78 healthy participants. A separate phase I trial in China, CTR20221542, also demonstrated comparable safety and pharmacokinetic profiles. This work was completed in roughly 18 months from target discovery to preclinical candidate nomination and demonstrates the capabilities of our generative AI-driven drug-discovery pipeline.

Defining the progeria phenome

Progeroid disorders are a heterogenous group of rare and complex hereditary syndromes presenting with pleiotropic phenotypes associated with normal aging. Due to the large variation in clinical presentation the diseases pose a diagnostic challenge for clinicians which consequently restricts medical research. To accommodate the challenge, we compiled a list of known progeroid syndromes and calculated the mean prevalence of their associated phenotypes, defining what we term the 'progeria phenome'. The data were used to train a support vector machine that is available at https://www.mitodb.com and able to classify progerias based on phenotypes. Furthermore, this allowed us to investigate the correlation of progeroid syndromes and syndromes with various pathogenesis using hierarchical clustering algorithms and disease networks. We detected that ataxia-telangiectasia like disorder 2, spastic paraplegia 49 and Meier-Gorlin syndrome display strong association to progeroid syndromes, thereby implying that the syndromes are previously unrecognized progerias. In conclusion, our study has provided tools to evaluate the likelihood of a syndrome or patient being progeroid. This is a considerable step forward in our understanding of what constitutes a premature aging disorder and how to diagnose them.

Transcriptome-based biomarker prediction for Parkinson's disease using genome-scale metabolic modeling

Parkinson's disease (PD) is the second most common neurodegenerative disease in the world. Identification of PD biomarkers is crucial for early diagnosis and to develop target-based therapeutic agents. Integrative analysis of genome-scale metabolic models (GEMs) and omics data provides a computational approach for the prediction of metabolite biomarkers. Here, we applied the TIMBR (Transcriptionally Inferred Metabolic Biomarker Response) algorithm and two modified versions of TIMBR to investigate potential metabolite biomarkers for PD. To this end, we mapped thirteen post-mortem PD transcriptome datasets from the substantia nigra region onto Human-GEM. We considered a metabolite as a candidate biomarker if its production was predicted to be more efficient by a TIMBR-family algorithm in control or PD case for the majority of the datasets. Different metrics based on well-known PD-related metabolite alterations, PD-associated pathways, and a list of 25 high-confidence PD metabolite biomarkers compiled from the literature were used to compare the prediction performance of the three algorithms tested. The modified algorithm with the highest prediction power based on the metrics was called TAMBOOR, TrAnscriptome-based Metabolite Biomarkers by On-Off Reactions, which was introduced for the first time in this study. TAMBOOR performed better in terms of capturing well-known pathway alterations and metabolite secretion changes in PD. Therefore, our tool has a strong potential to be used for the prediction of novel diagnostic biomarkers for human diseases.

Drug discovery and therapeutic perspectives for proximal tubulopathies

The efficient reabsorption of essential nutrients by epithelial cells in the proximal tubule of the kidney is crucial for maintaining homeostasis. This process relies heavily on a complex ecosystem of vesicular trafficking pathways. At the center of this network, the lysosome plays a pivotal role in processing incoming molecules, sensing nutrient availability, sorting receptors and transporters, and balancing differentiation and proliferation in the tubular epithelial cells. Disruptions in these fundamental processes can lead to proximal tubulopathy-a condition characterized by the dysfunction of the tubular cells followed by the presence of low-molecular-weight proteins and solutes in urine. If left untreated, proximal tubulopathy can progress to chronic kidney disease and severe complications. Functional studies of rare inherited disorders affecting the proximal tubule have gleaned actionable insights into fundamental mechanisms of homeostasis while revealing drug targets for therapeutic discovery and development. In this mini review, we explore hereditary proximal tubulopathies as a paradigm of kidney homeostasis disorders, discussing the factors contributing to tubular dysfunction. In addition, we shed light on the current landscape of drug discovery approaches used to identify actionable targets and summarize the preclinical pipeline of potential therapeutic agents. These efforts may ultimately lead to new treatment avenues for proximal tubulopathies, which are currently inadequately tackled by existing therapies. Through this article, our hope is to promote academia-industry partnerships and advocate for research consortia that can accelerate the effective translation of knowledge advances into innovative therapies addressing the huge unmet needs of individuals with these debilitating diseases.

A comprehensive AI-driven analysis of large-scale omic datasets reveals novel dual-purpose targets for the treatment of cancer and aging

As aging and tumorigenesis are tightly interconnected biological processes, targeting their common underlying driving pathways may induce dual-purpose anti-aging and anti-cancer effects. Our transcriptomic analyses of 16,740 healthy samples demonstrated tissue-specific age-associated gene expression, with most tumor suppressor genes downregulated during aging. Furthermore, a large-scale pan-cancer analysis of 11 solid tumor types (11,303 cases and 4431 control samples) revealed that many cellular processes, such as protein localization, DNA replication, DNA repair, cell cycle, and RNA metabolism, were upregulated in cancer but downregulated in healthy aging tissues, whereas pathways regulating cellular senescence were upregulated in both aging and cancer. Common cancer targets were identified by the AI-driven target discovery platform-PandaOmics. Age-associated cancer targets were selected and further classified into four groups based on their reported roles in lifespan. Among the 51 identified age-associated cancer targets with anti-aging experimental evidence, 22 were proposed as dual-purpose targets for anti-aging and anti-cancer treatment with the same therapeutic direction. Among age-associated cancer targets without known lifespan-regulating activity, 23 genes were selected based on predicted dual-purpose properties. Knockdown of histone demethylase KDM1A, one of these unexplored candidates, significantly extended lifespan in Caenorhabditis elegans. Given KDM1A's anti-cancer activities reported in both preclinical and clinical studies, our findings propose KDM1A as a promising dual-purpose target. This is the first study utilizing an innovative AI-driven approach to identify dual-purpose target candidates for anti-aging and anti-cancer treatment, supporting the value of AI-assisted target identification for drug discovery.

The promise of explainable deep learning for omics data analysis: Adding new discovery tools to AI

Deep learning has already revolutionised the way a wide range of data is processed in many areas of daily life. The ability to learn abstractions and relationships from heterogeneous data has provided impressively accurate prediction and classification tools to handle increasingly big datasets. This has a significant impact on the growing wealth of omics datasets, with the unprecedented opportunity for a better understanding of the complexity of living organisms. While this revolution is transforming the way these data are analyzed, explainable deep learning is emerging as an additional tool with the potential to change the way biological data is interpreted. Explainability addresses critical issues such as transparency, so important when computational tools are introduced especially in clinical environments. Moreover, it empowers artificial intelligence with the capability to provide new insights into the input data, thus adding an element of discovery to these already powerful resources. In this review, we provide an overview of the transformative effects explainable deep learning is having on multiple sectors, ranging from genome engineering and genomics, from radiomics to drug design and clinical trials. We offer a perspective to life scientists, to better understand the potential of these tools, and a motivation to implement them in their research, by suggesting learning resources they can use to move their first steps in this field.

Prediction of Clinical Trials Outcomes Based on Target Choice and Clinical Trial Design with Multi-Modal Artificial Intelligence

Drug discovery and development is a notoriously risky process with high failure rates at every stage, including disease modeling, target discovery, hit discovery, lead optimization, preclinical development, human safety, and efficacy studies. Accurate prediction of clinical trial outcomes may help significantly improve the efficiency of this process by prioritizing therapeutic programs that are more likely to succeed in clinical trials and ultimately benefit patients. Here, we describe inClinico, a transformer-based artificial intelligence software platform designed to predict the outcome of phase II clinical trials. The platform combines an ensemble of clinical trial outcome prediction engines that leverage generative artificial intelligence and multimodal data, including omics, text, clinical trial design, and small molecule properties. inClinico was validated in retrospective, quasi-prospective, and prospective validation studies internally and with pharmaceutical companies and financial institutions. The platform achieved 0.88 receiver operating characteristic area under the curve in predicting the phase II to phase III transition on a quasi-prospective validation dataset. The first prospective predictions were made and placed on date-stamped preprint servers in 2016. To validate our model in a real-world setting, we published forecasted outcomes for several phase II clinical trials achieving 79% accuracy for the trials that have read out. We also present an investment application of inClinico using date stamped virtual trading portfolio demonstrating 35% 9-month return on investment.

Multiomic prediction of therapeutic targets for human diseases associated with protein phase separation

The phenomenon of protein phase separation (PPS) underlies a wide range of cellular functions. Correspondingly, the dysregulation of the PPS process has been associated with numerous human diseases. To enable therapeutic interventions based on the regulation of this association, possible targets should be identified. For this purpose, we present an approach that combines the multiomic PandaOmics platform with the FuzDrop method to identify PPS-prone disease-associated proteins. Using this approach, we prioritize candidates with high PandaOmics and FuzDrop scores using a profiling method that accounts for a wide range of parameters relevant for disease mechanism and pharmacological intervention. We validate the differential phase separation behaviors of three predicted Alzheimer's disease targets (MARCKS, CAMKK2, and p62) in two cell models of this disease. Overall, the approach that we present generates a list of possible therapeutic targets for human diseases associated with the dysregulation of the PPS process.

Uniformly shaped harmonization combines human transcriptomic data from different platforms while retaining their biological properties and differential gene expression patterns

Introduction: Co-normalization of RNA profiles obtained using different experimental platforms and protocols opens avenue for comprehensive comparison of relevant features like differentially expressed genes associated with disease. Currently, most of bioinformatic tools enable normalization in a flexible format that depends on the individual datasets under analysis. Thus, the output data of such normalizations will be poorly compatible with each other. Recently we proposed a new approach to gene expression data normalization termed Shambhala which returns harmonized data in a uniform shape, where every expression profile is transformed into a pre-defined universal format. We previously showed that following shambhalization of human RNA profiles, overall tissue-specific clustering features are strongly retained while platform-specific clustering is dramatically reduced. Methods: Here, we tested Shambhala performance in retention of fold-change gene expression features and other functional characteristics of gene clusters such as pathway activation levels and predicted cancer drug activity scores. Results: Using 6,793 cancer and 11,135 normal tissue gene expression profiles from the literature and experimental datasets, we applied twelve performance criteria for different versions of Shambhala and other methods of transcriptomic harmonization with flexible output data format. Such criteria dealt with the biological type classifiers, hierarchical clustering, correlation/regression properties, stability of drug efficiency scores, and data quality for using machine learning classifiers. Discussion: Shambhala-2 harmonizer demonstrated the best results with the close to 1 correlation and linear regression coefficients for the comparison of training vs validation datasets and more than two times lesser instability for calculation of drug efficiency scores compared to other methods.

Algorithmically Reconstructed Molecular Pathways as the New Generation of Prognostic Molecular Biomarkers in Human Solid Cancers

Individual gene expression and molecular pathway activation profiles were shown to be effective biomarkers in many cancers. Here, we used the human interactome model to algorithmically build 7470 molecular pathways centered around individual gene products. We assessed their associations with tumor type and survival in comparison with the previous generation of molecular pathway biomarkers (3022 "classical" pathways) and with the RNA transcripts or proteomic profiles of individual genes, for 8141 and 1117 samples, respectively. For all analytes in RNA and proteomic data, respectively, we found a total of 7441 and 7343 potential biomarker associations for gene-centric pathways, 3020 and 2950 for classical pathways, and 24,349 and 6742 for individual genes. Overall, the percentage of RNA biomarkers was statistically significantly higher for both types of pathways than for individual genes (p < 0.05). In turn, both types of pathways showed comparable performance. The percentage of cancer-type-specific biomarkers was comparable between proteomic and transcriptomic levels, but the proportion of survival biomarkers was dramatically lower for proteomic data. Thus, we conclude that pathway activation level is the advanced type of biomarker for RNA and proteomic data, and momentary algorithmic computer building of pathways is a new credible alternative to time-consuming hypothesis-driven manual pathway curation and reconstruction.

Quantum computing for near-term applications in generative chemistry and drug discovery

In recent years, drug discovery and life sciences have been revolutionized with machine learning and artificial intelligence (AI) methods. Quantum computing is touted to be the next most significant leap in technology; one of the main early practical applications for quantum computing solutions is predicted to be in quantum chemistry simulations. Here, we review the near-term applications of quantum computing and their advantages for generative chemistry and highlight the challenges that can be addressed with noisy intermediate-scale quantum (NISQ) devices. We also discuss the possible integration of generative systems running on quantum computers into established generative AI platforms.

Lysosomal cystine export regulates mTORC1 signaling to guide kidney epithelial cell fate specialization

Differentiation is critical for cell fate decisions, but the signals involved remain unclear. The kidney proximal tubule (PT) cells reabsorb disulphide-rich proteins through endocytosis, generating cystine via lysosomal proteolysis. Here we report that defective cystine mobilization from lysosomes through cystinosin (CTNS), which is mutated in cystinosis, diverts PT cells towards growth and proliferation, disrupting their functions. Mechanistically, cystine storage stimulates Ragulator-Rag GTPase-dependent recruitment of mechanistic target of rapamycin complex 1 (mTORC1) and its constitutive activation. Re-introduction of CTNS restores nutrient-dependent regulation of mTORC1 in knockout cells, whereas cell-permeant analogues of L-cystine, accumulating within lysosomes, render wild-type cells resistant to nutrient withdrawal. Therapeutic mTORC1 inhibition corrects lysosome and differentiation downstream of cystine storage, and phenotypes in preclinical models of cystinosis. Thus, cystine serves as a lysosomal signal that tailors mTORC1 and metabolism to direct epithelial cell fate decisions. These results identify mechanisms and therapeutic targets for dysregulated homeostasis in cystinosis.

ProAct: quantifying the differential activity of biological processes in tissues, cells, and user-defined contexts

The distinct functions and phenotypes of human tissues and cells derive from the activity of biological processes that varies in a context-dependent manner. Here, we present the Process Activity (ProAct) webserver that estimates the preferential activity of biological processes in tissues, cells, and other contexts. Users can upload a differential gene expression matrix measured across contexts or cells, or use a built-in matrix of differential gene expression in 34 human tissues. Per context, ProAct associates gene ontology (GO) biological processes with estimated preferential activity scores, which are inferred from the input matrix. ProAct visualizes these scores across processes, contexts, and process-associated genes. ProAct also offers potential cell-type annotations for cell subsets, by inferring them from the preferential activity of 2001 cell-type-specific processes. Thus, ProAct output can highlight the distinct functions of tissues and cell types in various contexts, and can enhance cell-type annotation efforts. The ProAct webserver is available at https://netbio.bgu.ac.il/ProAct/.

Identification of dual-purpose therapeutic targets implicated in aging and glioblastoma multiforme using PandaOmics - an AI-enabled biological target discovery platform

Glioblastoma Multiforme (GBM) is the most aggressive and most common primary malignant brain tumor. The age of GBM patients is considered as one of the disease's negative prognostic factors and the mean age of diagnosis is 62 years. A promising approach to preventing both GBM and aging is to identify new potential therapeutic targets that are associated with both conditions as concurrent drivers. In this work, we present a multi-angled approach of identifying targets, which takes into account not only the disease-related genes but also the ones important in aging. For this purpose, we developed three strategies of target identification using the results of correlation analysis augmented with survival data, differences in expression levels and previously published information of aging-related genes. Several studies have recently validated the robustness and applicability of AI-driven computational methods for target identification in both cancer and aging-related diseases. Therefore, we leveraged the AI predictive power of the PandaOmics TargetID engine in order to rank the resulting target hypotheses and prioritize the most promising therapeutic gene targets. We propose cyclic nucleotide gated channel subunit alpha 3 (CNGA3), glutamate dehydrogenase 1 (GLUD1) and sirtuin 1 (SIRT1) as potential novel dual-purpose therapeutic targets to treat aging and GBM.

AlphaFold accelerates artificial intelligence powered drug discovery: efficient discovery of a novel CDK20 small molecule inhibitor

The application of artificial intelligence (AI) has been considered a revolutionary change in drug discovery and development. In 2020, the AlphaFold computer program predicted protein structures for the whole human genome, which has been considered a remarkable breakthrough in both AI applications and structural biology. Despite the varying confidence levels, these predicted structures could still significantly contribute to structure-based drug design of novel targets, especially the ones with no or limited structural information. In this work, we successfully applied AlphaFold to our end-to-end AI-powered drug discovery engines, including a biocomputational platform PandaOmics and a generative chemistry platform Chemistry42. A novel hit molecule against a novel target without an experimental structure was identified, starting from target selection towards hit identification, in a cost- and time-efficient manner. PandaOmics provided the protein of interest for the treatment of hepatocellular carcinoma (HCC) and Chemistry42 generated the molecules based on the structure predicted by AlphaFold, and the selected molecules were synthesized and tested in biological assays. Through this approach, we identified a small molecule hit compound for cyclin-dependent kinase 20 (CDK20) with a binding constant Kd value of 9.2 ± 0.5 μM (n = 3) within 30 days from target selection and after only synthesizing 7 compounds. Based on the available data, a second round of AI-powered compound generation was conducted and through this, a more potent hit molecule, ISM042-2-048, was discovered with an average Kd value of 566.7 ± 256.2 nM (n = 3). Compound ISM042-2-048 also showed good CDK20 inhibitory activity with an IC50 value of 33.4 ± 22.6 nM (n = 3). In addition, ISM042-2-048 demonstrated selective anti-proliferation activity in an HCC cell line with CDK20 overexpression, Huh7, with an IC50 of 208.7 ± 3.3 nM, compared to a counter screen cell line HEK293 (IC50 = 1706.7 ± 670.0 nM). This work is the first demonstration of applying AlphaFold to the hit identification process in drug discovery.

Chemistry42: An AI-Driven Platform for Molecular Design and Optimization

Chemistry42 is a software platform for de novo small molecule design and optimization that integrates Artificial Intelligence (AI) techniques with computational and medicinal chemistry methodologies. Chemistry42 efficiently generates novel molecular structures with optimized properties validated in both in vitro and in vivo studies and is available through licensing or collaboration. Chemistry42 is the core component of Insilico Medicine's Pharma.ai drug discovery suite. Pharma.ai also includes PandaOmics for target discovery and multiomics data analysis, and inClinico─a data-driven multimodal forecast of a clinical trial's probability of success (PoS). In this paper, we demonstrate how the platform can be used to efficiently find novel molecular structures against DDR1 and CDK20.

Nanoparticles and convergence of artificial intelligence for targeted drug delivery for cancer therapy: Current progress and challenges

Cancer is a life-threatening disease, resulting in nearly 10 million deaths worldwide. There are various causes of cancer, and the prognostic information varies in each patient because of unique molecular signatures in the human body. However, genetic heterogeneity occurs due to different cancer types and changes in the neoplasms, which complicates the diagnosis and treatment. Targeted drug delivery is considered a pivotal contributor to precision medicine for cancer treatments as this method helps deliver medication to patients by systematically increasing the drug concentration on the targeted body parts. In such cases, nanoparticle-mediated drug delivery and the integration of artificial intelligence (AI) can help bridge the gap and enhance localized drug delivery systems capable of biomarker sensing. Diagnostic assays using nanoparticles (NPs) enable biomarker identification by accumulating in the specific cancer sites and ensuring accurate drug delivery planning. Integrating NPs for cancer targeting and AI can help devise sophisticated systems that further classify cancer types and understand complex disease patterns. Advanced AI algorithms can also help in biomarker detection, predicting different NP interactions of the targeted drug, and evaluating drug efficacy. Considering the advantages of the convergence of NPs and AI for targeted drug delivery, there has been significantly limited research focusing on the specific research theme, with most of the research being proposed on AI and drug discovery. Thus, the study's primary objective is to highlight the recent advances in drug delivery using NPs, and their impact on personalized treatment plans for cancer patients. In addition, a focal point of the study is also to highlight how integrating AI, and NPs can help address some of the existing challenges in drug delivery by conducting a collective survey.

High-confidence cancer patient stratification through multiomics investigation of DNA repair disorders

Multiple cancer types have limited targeted therapeutic options, in part due to incomplete understanding of the molecular processes underlying tumorigenesis and significant intra- and inter-tumor heterogeneity. Identification of novel molecular biomarkers stratifying cancer patients with different survival outcomes may provide new opportunities for target discovery and subsequent development of tailored therapies. Here, we applied the artificial intelligence-driven PandaOmics platform ( https://pandaomics.com/ ) to explore gene expression changes in rare DNA repair-deficient disorders and identify novel cancer targets. Our analysis revealed that CEP135, a scaffolding protein associated with early centriole biogenesis, is commonly downregulated in DNA repair diseases with high cancer predisposition. Further screening of survival data in 33 cancers available at TCGA database identified sarcoma as a cancer type where lower survival was significantly associated with high CEP135 expression. Stratification of cancer patients based on CEP135 expression enabled us to examine therapeutic targets that could be used for the improvement of existing therapies against sarcoma. The latter was based on application of the PandaOmics target-ID algorithm coupled with in vitro studies that revealed polo-like kinase 1 (PLK1) as a potential therapeutic candidate in sarcoma patients with high CEP135 levels and poor survival. While further target validation is required, this study demonstrated the potential of in silico-based studies for a rapid biomarker discovery and target characterization.

Omics Data and Data Representations for Deep Learning-Based Predictive Modeling

Medical discoveries mainly depend on the capability to process and analyze biological datasets, which inundate the scientific community and are still expanding as the cost of next-generation sequencing technologies is decreasing. Deep learning (DL) is a viable method to exploit this massive data stream since it has advanced quickly with there being successive innovations. However, an obstacle to scientific progress emerges: the difficulty of applying DL to biology, and this because both fields are evolving at a breakneck pace, thus making it hard for an individual to occupy the front lines of both of them. This paper aims to bridge the gap and help computer scientists bring their valuable expertise into the life sciences. This work provides an overview of the most common types of biological data and data representations that are used to train DL models, with additional information on the models themselves and the various tasks that are being tackled. This is the essential information a DL expert with no background in biology needs in order to participate in DL-based research projects in biomedicine, biotechnology, and drug discovery. Alternatively, this study could be also useful to researchers in biology to understand and utilize the power of DL to gain better insights into and extract important information from the omics data.

Identification of Therapeutic Targets for Amyotrophic Lateral Sclerosis Using PandaOmics – An AI-Enabled Biological Target Discovery Platform

Amyotrophic lateral sclerosis (ALS) is a severe neurodegenerative disease with ill-defined pathogenesis, calling for urgent developments of new therapeutic regimens. Herein, we applied PandaOmics, an AI-driven target discovery platform, to analyze the expression profiles of central nervous system (CNS) samples (237 cases; 91 controls) from public datasets, and direct iPSC-derived motor neurons (diMNs) (135 cases; 31 controls) from Answer ALS. Seventeen high-confidence and eleven novel therapeutic targets were identified and will be released onto ALS.AI (http://als.ai/). Among the proposed targets screened in the c9ALS Drosophila model, we verified 8 unreported genes (KCNB2, KCNS3, ADRA2B, NR3C1, P2RY14, PPP3CB, PTPRC, and RARA) whose suppression strongly rescues eye neurodegeneration. Dysregulated pathways identified from CNS and diMN data characterize different stages of disease development. Altogether, our study provides new insights into ALS pathophysiology and demonstrates how AI speeds up the target discovery process, and opens up new opportunities for therapeutic interventions.

Artificial intelligence in cancer target identification and drug discovery

Artificial intelligence is an advanced method to identify novel anticancer targets and discover novel drugs from biology networks because the networks can effectively preserve and quantify the interaction between components of cell systems underlying human diseases such as cancer. Here, we review and discuss how to employ artificial intelligence approaches to identify novel anticancer targets and discover drugs. First, we describe the scope of artificial intelligence biology analysis for novel anticancer target investigations. Second, we review and discuss the basic principles and theory of commonly used network-based and machine learning-based artificial intelligence algorithms. Finally, we showcase the applications of artificial intelligence approaches in cancer target identification and drug discovery. Taken together, the artificial intelligence models have provided us with a quantitative framework to study the relationship between network characteristics and cancer, thereby leading to the identification of potential anticancer targets and the discovery of novel drug candidates.

Transcriptomic Portraits and Molecular Pathway Activation Features of Adult Spinal Intramedullary Astrocytomas

In this study, we report 31 spinal intramedullary astrocytoma (SIA) RNA sequencing (RNA-seq) profiles for 25 adult patients with documented clinical annotations. To our knowledge, this is the first clinically annotated RNA-seq dataset of spinal astrocytomas derived from the intradural intramedullary compartment. We compared these tumor profiles with the previous healthy central nervous system (CNS) RNA-seq data for spinal cord and brain and identified SIA-specific gene sets and molecular pathways. Our findings suggest a trend for SIA-upregulated pathways governing interactions with the immune cells and downregulated pathways for the neuronal functioning in the context of normal CNS activity. In two patient tumor biosamples, we identified diagnostic KIAA1549-BRAF fusion oncogenes, and we also found 16 new SIA-associated fusion transcripts. In addition, we bioinformatically simulated activities of targeted cancer drugs in SIA samples and predicted that several tyrosine kinase inhibitory drugs and thalidomide analogs could be potentially effective as second-line treatment agents to aid in the prevention of SIA recurrence and progression.

Hallmarks of aging-based dual-purpose disease and age-associated targets predicted using PandaOmics AI-powered discovery engine

Aging biology is a promising and burgeoning research area that can yield dual-purpose pathways and protein targets that may impact multiple diseases, while retarding or possibly even reversing age-associated processes. One widely used approach to classify a multiplicity of mechanisms driving the aging process is the hallmarks of aging. In addition to the classic nine hallmarks of aging, processes such as extracellular matrix stiffness, chronic inflammation and activation of retrotransposons are also often considered, given their strong association with aging. In this study, we used a variety of target identification and prioritization techniques offered by the AI-powered PandaOmics platform, to propose a list of promising novel aging-associated targets that may be used for drug discovery. We also propose a list of more classical targets that may be used for drug repurposing within each hallmark of aging. Most of the top targets generated by this comprehensive analysis play a role in inflammation and extracellular matrix stiffness, highlighting the relevance of these processes as therapeutic targets in aging and age-related diseases. Overall, our study reveals both high confidence and novel targets associated with multiple hallmarks of aging and demonstrates application of the PandaOmics platform to target discovery across multiple disease areas.

Treating Duchenne Muscular Dystrophy: The Promise of Stem Cells, Artificial Intelligence, and Multi-Omics

Muscular dystrophies are chronic and debilitating disorders caused by progressive muscle wasting. Duchenne muscular dystrophy (DMD) is the most common type. DMD is a well-characterized genetic disorder caused by the absence of dystrophin. Although some therapies exist to treat the symptoms and there are ongoing efforts to correct the underlying molecular defect, patients with muscular dystrophies would greatly benefit from new therapies that target the specific pathways contributing directly to the muscle disorders. Three new advances are poised to change the landscape of therapies for muscular dystrophies such as DMD. First, the advent of human induced pluripotent stem cells (iPSCs) allows researchers to design effective treatment strategies that make up for the gaps missed by conventional “one size fits all” strategies. By characterizing tissue alterations with single-cell resolution and having molecular profiles for therapeutic treatments for a variety of cell types, clinical researchers can design multi-pronged interventions to not just delay degenerative processes, but regenerate healthy tissues. Second, artificial intelligence (AI) will play a significant role in developing future therapies by allowing the aggregation and synthesis of large and disparate datasets to help reveal underlying molecular mechanisms. Third, disease models using a high volume of multi-omics data gathered from diverse sources carry valuable information about converging and diverging pathways. Using these new tools, the results of previous and emerging studies will catalyze precision medicine-based drug development that can tackle devastating disorders such as DMD.

The differential activity of biological processes in tissues and cell subsets can illuminate disease-related processes and cell-type identities

MOTIVATION

The distinct functionalities of human tissues and cell types underlie complex phenotype-genotype relationships, yet often remain elusive. Harnessing the multitude of bulk and single-cell human transcriptomes while focusing on processes can help reveal these distinct functionalities.

RESULTS

The Tissue-Process Activity (TiPA) method aims to identify processes that are preferentially active or under-expressed in specific contexts, by comparing the expression levels of process genes between contexts. We tested TiPA on 1579 tissue-specific processes and bulk tissue transcriptomes, finding that it performed better than another method. Next, we used TiPA to ask whether the activity of certain processes could underlie the tissue-specific manifestation of 1233 hereditary diseases. We found that 21% of the disease-causing genes indeed participated in such processes, thereby illuminating their genotype-phenotype relationships. Lastly, we applied TiPA to single-cell transcriptomes of 108 human cell types, revealing that process activities often match cell-type identities and can thus aid annotation efforts. Hence, differential activity of processes can highlight the distinct functionality of tissues and cells in a robust and meaningful manner.

AVAILABILITY AND IMPLEMENTATION

TiPA code is available in GitHub (https://github.com/moranshar/TiPA). In addition, all data are available as part of the Supplementary Material.

SUPPLEMENTARY INFORMATION

Supplementary data are available at Bioinformatics online.

Meeting Report: Aging Research and Drug Discovery

Aging is the single largest risk factor for most chronic diseases, and thus possesses large socioeconomic interest to continuously aging societies. Consequently, the field of aging research is expanding alongside a growing focus from the industry and investors in aging research. This year’s 8th Annual Aging Research and Drug Discovery (ARDD) meeting was organized as a hybrid meeting from August 30th to September 3rd 2021 with more than 130 attendees participating on-site at the Ceremonial Hall at University of Copenhagen, Denmark, and 1800 engaging online. The conference comprised of presentations from 75 speakers focusing on new research in topics including mechanisms of aging and how these can be modulated as well as the use of AI and new standards of practices within aging research. This year, a longevity workshop was included to build stronger connections with the clinical community.

Identifying Molecular Biomarkers for Diseases With Machine Learning Based on Integrative Omics

Molecular biomarkers are certain molecules or set of molecules that can be of help for diagnosis or prognosis of diseases or disorders. In the past decades, thanks to the advances in high-throughput technologies, a huge amount of molecular 'omics' data, e.g., transcriptomics and proteomics, have been accumulated. The availability of these omics data makes it possible to screen biomarkers for diseases or disorders. Accordingly, a number of computational approaches have been developed to identify biomarkers by exploring the omics data. In this review, we present a comprehensive survey on the recent progress of identification of molecular biomarkers with machine learning approaches. Specifically, we categorize the machine learning approaches into supervised, un-supervised and recommendation approaches, where the biomarkers including single genes, gene sets and small gene networks. In addition, we further discuss potential problems underlying bio-medical data that may pose challenges for machine learning, and provide possible directions for future biomarker identification.

Machine learning applications for therapeutic tasks with genomics data

Thanks to the increasing availability of genomics and other biomedical data, many machine learning algorithms have been proposed for a wide range of therapeutic discovery and development tasks. In this survey, we review the literature on machine learning applications for genomics through the lens of therapeutic development. We investigate the interplay among genomics, compounds, proteins, electronic health records, cellular images, and clinical texts. We identify 22 machine learning in genomics applications that span the whole therapeutics pipeline, from discovering novel targets, personalizing medicine, developing gene-editing tools, all the way to facilitating clinical trials and post-market studies. We also pinpoint seven key challenges in this field with potentials for expansion and impact. This survey examines recent research at the intersection of machine learning, genomics, and therapeutic development.

Effector T cell responses unleashed by regulatory T cell ablation exacerbate oral squamous cell carcinoma

Immune suppression by CD4+FOXP3+ regulatory T (Treg) cells and tumor infiltration by CD8+ effector T cells represent two major factors impacting response to cancer immunotherapy. Using deconvolution-based transcriptional profiling of human papilloma virus (HPV)-negative oral squamous cell carcinomas (OSCCs) and other solid cancers, we demonstrate that the density of Treg cells does not correlate with that of CD8+ T cells in many tumors, revealing polarized clusters enriched for either CD8+ T cells or CD4+ Treg and conventional T cells. In a mouse model of carcinogen-induced OSCC characterized by CD4+ T cell enrichment, late-stage Treg cell ablation triggers increased densities of both CD4+ and CD8+ effector T cells within oral lesions. Notably, this intervention does not induce tumor regression but instead induces rapid emergence of invasive OSCCs via an effector T cell-dependent process. Thus, induction of a T cell-inflamed phenotype via therapeutic manipulation of Treg cells may trigger unexpected tumor-promoting effects in OSCC.

TGFBR2 mutation predicts resistance to immune checkpoint inhibitors in patients with non-small cell lung cancer

Background:

Resistance or even hyper-progression to immune checkpoint inhibitors (ICIs) manifesting as accelerated disease progression or death has impeded the clinical use of ICIs. The transforming growth factor beta (TGFβ) receptor pathway has been identified in contributing to immune dysfunction, which might be associated with resistance to ICIs. We aimed to explore the role of TGFβ in the resistance to ICIs in non-small cell lung cancer (NSCLC) in this study.

Methods:

Public cohorts with patients treated with ICIs or chemotherapy including POPLAR/OAK (n = 853), MSKCC (n = 1662) and Van Allen (n = 57) and TCGA (n = 3210) cohorts were obtained and analyzed.

Results:

The expression of immune-checkpoint related genes, including programmed death-ligand 1 (CD274), lymphocyte-activation gene 3 (LAG3), T cell immunoreceptor with Ig and ITIM domains (TIGIT), cytotoxic T-lymphocyte-associated protein 4 (CTLA-4), programmed cell death ligand 1 (PDCD1), and programmed cell death 1 ligand 2 (PDCD1LG2) were significantly upregulated in transforming growth factor beta TGFβ receptor 2 (TGFβR2)-mutated patients than those with wild-type TGFBR2 (p < 0.05). In the POPLAR/OAK cohort, TGFBR2-mutated patients showed shorter progression-free survival (PFS) [ p = 0.004; hazard ratio (HR), 2.83; 95% confidence interval (CI), 1.34–6.00] and overall survival (OS) ( p = 0.0006; HR, 3.46; 95% CI, 1.63–7.35) than those with wild-type TGFBR2 when treated with ICIs but not chemotherapy. In the merged MSKCC and Van Allen cohorts, a similar result was observed that the OS was inferior in patients with mutated TGFBR2 compared with those with wild-type TGFBR2 (p = 0.007; HR, 2.53; 95% CI, 1.25–5.12). The association between TGBFR2 mutation and survival remained significant in multivariable cox regression in both POPLAR/OAK cohort (p = 0.02; HR, 2.53; 95% CI, 1.17–5.45) and merged cohort (p = 0.008; HR, 2.63; 95% CI, 1.29–5.35). We further evaluated the association between TGFBR2 mutations and OS in multiple types of tumors. The association between TGFBR2 mutations and OS remained significant in NSCLC (p = 0.02; HR, 2.47; 95% CI, 1.16–5.26), but not in other type of tumors.

Conclusions:

We identified that TGFBR2 mutation predicted the resistance to ICIs in NSCLCs. The clinical delivery of ICIs should be cautious in those patients.

Doublecortin-Like Kinase 1 (DCLK1) Is a Novel NOTCH Pathway Signaling Regulator in Head and Neck Squamous Cell Carcinoma

Despite recent advancements, the 5 year survival of head and neck squamous cell carcinoma (HNSCC) hovers at 60%. DCLK1 has been shown to regulate epithelial-to-mesenchymal transition as well as serving as a cancer stem cell marker in colon, pancreatic and renal cancer. Although it was reported that DCLK1 is associated with poor prognosis in oropharyngeal cancers, very little is known about the molecular characterization of DCLK1 in HNSCC. In this study, we performed a comprehensive transcriptome-based computational analysis on hundreds of HNSCC patients from TCGA and GEO databases, and found that DCLK1 expression positively correlates with NOTCH signaling pathway activation. Since NOTCH signaling has a recognized role in HNSCC tumorigenesis, we next performed a series of in vitro experiments in a collection of HNSCC cell lines to investigate the role of DCLK1 in NOTCH pathway regulation. Our analyses revealed that DCLK1 inhibition, using either a pharmacological inhibitor or siRNA, resulted in substantially decreased proliferation, invasion, migration, and colony formation. Furthermore, these effects paralleled downregulation of active NOTCH1, and its downstream effectors, HEY1, HES1 and HES5, whereas overexpression of DCLK1 in normal keratinocytes, lead to an upregulation of NOTCH signaling associated with increased proliferation. Analysis of 233 primary and 40 recurrent HNSCC cancer biopsies revealed that high DCLK1 expression was associated with poor prognosis and showed a trend towards higher active NOTCH1 expression in tumors with elevated DCLK1. Our results demonstrate the novel role of DCLK1 as a regulator of NOTCH signaling network and suggest its potential as a therapeutic target in HNSCC.

Artificial Intelligence in Pharmaceutical Field - A Critical Review

Artificial intelligence is an emerging sector in almost all fields. It is not confined only to a particular category and can be used in various fields like research, technology, and health. AI mainly concentrates on how computers analyze data and mimic the human thought process. As drug development involves high R & D costs and uncertainty in time consumption, artificial intelligence can serve as one of the promising solutions to overcome all these demerits. Due to the availability of enormous data, there are chances of missing out on some crucial details. For solving these issues, algorithms like machine learning, deep learning, and other expert systems are being used. On successful implementation of AI in the pharmaceutical field, the delays in drug development, and failure at the clinical and marketing level can be reduced. This review comprises information regarding the development of AI, its subfields, its overall implementation, and its application in the pharmaceutical sector and provides insights on challenges and limitations concerning AI.

Using proteomic and transcriptomic data to assess activation of intracellular molecular pathways

Analysis of molecular pathway activation is the recent instrument that helps to quantize activities of various intracellular signaling, structural, DNA synthesis and repair, and biochemical processes. This may have a deep impact in fundamental research, bioindustry, and medicine. Unlike gene ontology analyses and numerous qualitative methods that can establish whether a pathway is affected in principle, the quantitative approach has the advantage of exactly measuring the extent of a pathway up/downregulation. This results in emergence of a new generation of molecular biomarkers-pathway activation levels, which reflect concentration changes of all measurable pathway components. The input data can be the high-throughput proteomic or transcriptomic profiles, and the output numbers take both positive and negative values and positively reflect overall pathway activation. Due to their nature, the pathway activation levels are more robust biomarkers compared to the individual gene products/protein levels. Here, we review the current knowledge of the quantitative gene expression interrogation methods and their applications for the molecular pathway quantization. We consider enclosed bioinformatic algorithms and their applications for solving real-world problems. Besides a plethora of applications in basic life sciences, the quantitative pathway analysis can improve molecular design and clinical investigations in pharmaceutical industry, can help finding new active biotechnological components and can significantly contribute to the progressive evolution of personalized medicine. In addition to the theoretical principles and concepts, we also propose publicly available software for the use of large-scale protein/RNA expression data to assess the human pathway activation levels.

Systematic investigation of mouse models of Parkinson's disease by transcriptome mapping on a brain-specific genome-scale metabolic network

Genome-scale metabolic networks enable systemic investigation of metabolic alterations caused by diseases by providing interpretation of omics data. Although Mus musculus (mouse) is one of the most commonly used model organisms for neurodegenerative diseases, a brain-specific metabolic network model of mice has not yet been reconstructed. Here we reconstructed the first brain-specific metabolic network model of mice, iBrain674-Mm, by a homology-based approach, which consisted of 992 reactions controlled by 674 genes and distributed over 48 pathways. We validated the newly reconstructed network model by showing that it predicts healthy resting-state metabolic phenotypes of mouse brain compatible with the literature. We later used iBrain674-Mm to interpret various experimental mouse models of Parkinson's Disease (PD) at the transcriptome level. To this end, we applied a constraint-based modelling based biomarker prediction method called TIMBR (Transcriptionally Inferred Metabolic Biomarker Response) to predict altered metabolite production from transcriptomic data. Systemic analysis of seven different PD mouse models by TIMBR showed that the neuronal levels of glutamate, lactate, creatine phosphate, neuronal acetylcholine, bilirubin and formate increased in most of the PD mouse models, whereas the levels of melatonin, epinephrine, astrocytic formate and astrocytic bilirubin decreased. Although most of the predictions were consistent with the literature, there were some inconsistencies among different PD mouse models, signifying that there is no perfect experimental model to reflect PD metabolism. The newly reconstructed brain-specific genome-scale metabolic network model of mice can make important contributions to the interpretation and development of experimental mouse models of PD and other neurodegenerative diseases.

Human blood serum can donor-specifically antagonize effects of EGFR-targeted drugs on squamous carcinoma cell growth

Many patients fail to respond to EGFR-targeted therapeutics, and personalized diagnostics is needed to identify putative responders. We investigated 1630 colorectal and lung squamous carcinomas and 1357 normal lung and colon samples and observed huge variation in EGFR pathway activation in both cancerous and healthy tissues, irrespectively on EGFR gene mutation status. We investigated whether human blood serum can affect squamous carcinoma cell growth and EGFR drug response. We demonstrate that human serum antagonizes the effects of EGFR-targeted drugs erlotinib and cetuximab on A431 squamous carcinoma cells by increasing IC50 by about 2- and 20-fold, respectively. The effects on clonogenicity varied significantly across the individual serum samples in every experiment, with up to 100% differences. EGF concentration could explain many effects of blood serum samples, and EGFR ligands-depleted serum showed lesser effect on drug sensitivity.

Algorithmic Annotation of Functional Roles for Components of 3,044 Human Molecular Pathways

Current methods of high-throughput molecular and genomic analyses enabled to reconstruct thousands of human molecular pathways. Knowledge of molecular pathways structure and architecture taken along with the gene expression data can help interrogating the pathway activation levels (PALs) using different bioinformatic algorithms. In turn, the pathway activation profiles can characterize molecular processes, which are differentially regulated and give numeric characteristics of the extent of their activation or inhibition. However, different pathway nodes may have different functions toward overall pathway regulation, and calculation of PAL requires knowledge of molecular function of every node in the pathway in terms of its activator or inhibitory role. Thus, high-throughput annotation of functional roles of pathway nodes is required for the comprehensive analysis of the pathway activation profiles. We proposed an algorithm that identifies functional roles of the pathway components and applied it to annotate 3,044 human molecular pathways extracted from the Biocarta, Reactome, KEGG, Qiagen Pathway Central, NCI, and HumanCYC databases and including 9,022 gene products. The resulting knowledgebase can be applied for the direct calculation of the PALs and establishing large scale profiles of the signaling, metabolic, and DNA repair pathway regulation using high throughput gene expression data. We also provide a bioinformatic tool for PAL data calculations using the current pathway knowledgebase.

System, Method and Software for Calculation of a Cannabis Drug Efficiency Index for the Reduction of Inflammation

There are many varieties of Cannabis sativa that differ from each other by composition of cannabinoids, terpenes and other molecules. The medicinal properties of these cultivars are often very different, with some being more efficient than others. This report describes the development of a method and software for the analysis of the efficiency of various cannabis extracts to detect the anti-inflammatory properties of the various cannabis extracts. The method uses high-throughput gene expression profiling data but can potentially use other omics data as well. According to the signaling pathway topology, the gene expression profiles are convoluted into the signaling pathway activities using a signaling pathway impact analysis (SPIA) method. The method was tested by inducing inflammation in human 3D epithelial tissues, including intestine, oral and skin, and then exposing these tissues to various extracts and then performing transcriptome analysis. The analysis showed a different efficiency of the various extracts in restoring the transcriptome changes to the pre-inflammation state, thus allowing to calculate a different cannabis drug efficiency index (CDEI).

Cancer gene expression profiles associated with clinical outcomes to chemotherapy treatments

Background

Machine learning (ML) methods still have limited applicability in personalized oncology due to low numbers of available clinically annotated molecular profiles. This doesn’t allow sufficient training of ML classifiers that could be used for improving molecular diagnostics.

Methods

We reviewed published datasets of high throughput gene expression profiles corresponding to cancer patients with known responses on chemotherapy treatments. We browsed Gene Expression Omnibus (GEO), The Cancer Genome Atlas (TCGA) and Tumor Alterations Relevant for GEnomics-driven Therapy (TARGET) repositories.

Results

We identified data collections suitable to build ML models for predicting responses on certain chemotherapeutic schemes. We identified 26 datasets, ranging from 41 till 508 cases per dataset. All the datasets identified were checked for ML applicability and robustness with leave-one-out cross validation. Twenty-three datasets were found suitable for using ML that had balanced numbers of treatment responder and non-responder cases.

Conclusions

We collected a database of gene expression profiles associated with clinical responses on chemotherapy for 2786 individual cancer cases. Among them seven datasets included RNA sequencing data (for 645 cases) and the others – microarray expression profiles. The cases represented breast cancer, lung cancer, low-grade glioma, endothelial carcinoma, multiple myeloma, adult leukemia, pediatric leukemia and kidney tumors. Chemotherapeutics included taxanes, bortezomib, vincristine, trastuzumab, letrozole, tipifarnib, temozolomide, busulfan and cyclophosphamide.

Radioprotectors.org: an open database of known and predicted radioprotectors

The search for radioprotectors is an ambitious goal with many practical applications. Particularly, the improvement of human radioresistance for space is an important task, which comes into view with the recent successes in the space industry. Currently, all radioprotective drugs can be divided into two large groups differing in their effectiveness depending on the type of exposure. The first of these is radioprotectors, highly effective for pulsed, and some types of relatively short exposure to irradiation. The second group consists of long-acting radioprotectors. These drugs are effective for prolonged and fractionated irradiation. They also protect against impulse exposure to ionizing radiation, but to a lesser extent than short-acting radioprotectors. Creating a database on radioprotectors is a necessity dictated by the modern development of science and technology. We have created an open database, Radioprotectors.org, containing an up-to-date list of substances with proven radioprotective properties. All radioprotectors are annotated with relevant chemical and biological information, including transcriptomic data, and can be filtered according to their properties. Additionally, the performed transcriptomics analysis has revealed specific transcriptomic profiles of radioprotectors, which should facilitate the search for potent radioprotectors.

Predictors of Neoadjuvant Chemotherapy Response in Breast Cancer: A Review

Neoadjuvant chemotherapy (NAC) largely increases operative chances and improves prognosis of the local advanced breast cancer patients. However, no specific means have been invented to predict the therapy responses of patients receiving NAC. Therefore, we focus on the alterations of tumor tissue-related microenvironments such as stromal tumor-infiltrating lymphocytes status, cyclin-dependent kinase expression, non-coding RNA transcription or other small molecular changes, in order to detect potentially predicted biomarkers which reflect the therapeutic efficacy of NAC in different subtypes of breast cancer. Further, possible mechanisms are also discussed to discover feasible treatment targets. Thus, these findings will be helpful to promote the prognosis of breast cancer patients who received NAC and summarized in this review.

A novel drug repurposing approach for non-small cell lung cancer using deep learning

Drug repurposing is an attractive and pragmatic way offering reduced risks and development time in the complicated process of drug discovery. In the past, drug repurposing has been largely accidental and serendipitous. The most successful examples so far have not involved a systematic approach. Nowadays, remarkable advances in drugs, diseases and bioinformatic knowledge are offering great opportunities for designing novel drug repurposing approach through comprehensive understanding of drug information. In this study, we introduced a novel drug repurposing approach based on transcriptomic data and chemical structures using deep learning. One strong candidate for repurposing has been identified. Pimozide is an anti-dyskinesia agent that is used for the suppression of motor and phonic tics in patients with Tourette's Disorder. However, our pipeline proposed it as a strong candidate for treating non-small cell lung cancer. The cytotoxicity of pimozide against A549 cell lines has been validated.

Replicative and radiation-induced aging: a comparison of gene expression profiles

All living organisms are subject to the aging process and experience the effect of ionizing radiation throughout their life. There have been a number of studies that linked ionizing radiation process to accelerated aging, but comprehensive signalome analysis of both processes was rarely conducted. Here we present a comparative signaling pathway based analysis of the transcriptomes of fibroblasts irradiated with different doses of ionizing radiation, replicatively aged fibroblasts and fibroblasts collected from young, middle age and old patients. We demonstrate a significant concordance between irradiation-induced and replicative senescence signalome signatures of fibroblasts. Additionally, significant differences in transcriptional response were also observed between fibroblasts irradiated with high and low dose. Our data shows that the transcriptome of replicatively aged fibroblasts is more similar to the transcriptome of the cells irradiated with 2 Gy, than with 5 сGy.This work revealed a number of signaling pathways that are shared between senescence and irradiation processes and can potentially be targeted by the new generation of gero- and radioprotectors.

BayesENproteomics: Bayesian Elastic Nets for Quantification of Peptidoforms in Complex Samples

Multivariate regression modelling provides a statistically powerful means of quantifying the effects of a given treatment while compensating for sources of variation and noise, such as variability between human donors and the behavior of different peptides during mass spectrometry. However, methods to quantify endogenous post-translational modifications (PTMs) are typically reliant on summary statistical methods that fail to consider sources of variability such as changes in the levels of the parent protein. Here, we compare three multivariate regression methods, including a novel Bayesian elastic net algorithm (BayesENproteomics) that enables assessment of relative protein abundances while also quantifying identified PTMs for each protein. We tested the ability of these methods to accurately quantify expression of proteins in a mixed-species benchmark experiment and to quantify synthetic PTMs induced by stable isotope labelling. Finally, we extended our regression pipeline to calculate fold changes at the pathway level, providing a complement to commonly used enrichment analysis. Our results show that BayesENproteomics can quantify changes to protein levels across a broad dynamic range while also accurately quantifying PTM and pathway-level fold changes.

Molecular Generation for Desired Transcriptome Changes With Adversarial Autoencoders

Gene expression profiles are useful for assessing the efficacy and side effects of drugs. In this paper, we propose a new generative model that infers drug molecules that could induce a desired change in gene expression. Our model-the Bidirectional Adversarial Autoencoder-explicitly separates cellular processes captured in gene expression changes into two feature sets: those related and unrelated to the drug incubation. The model uses related features to produce a drug hypothesis. We have validated our model on the LINCS L1000 dataset by generating molecular structures in the SMILES format for the desired transcriptional response. In the experiments, we have shown that the proposed model can generate novel molecular structures that could induce a given gene expression change or predict a gene expression difference after incubation of a given molecular structure. The code of the model is available at https://github.com/insilicomedicine/BiAAE.

RNA sequencing profiles and diagnostic signatures linked with response to ramucirumab in gastric cancer

Gastric cancer (GC) is the fifth-ranked cancer type by associated mortality. The proportion of early diagnosis is low, and most patients are diagnosed at the advanced stages. First-line therapy standardly includes fluoropyrimidines and platinum compounds with trastuzumab for HER2-positive cases. For recurrent disease, there are several alternative options including ramucirumab, a monoclonal therapeutic antibody that inhibits VEGF-mediated tumor angiogenesis by binding with VEGFR2, alone or in combination with other cancer drugs. However, overall response rate following ramucirumab or its combinations is 30%-80% of the patients, suggesting that personalization of drug prescription is needed to increase efficacy of treatment. We report here original tumor RNA sequencing profiles for 15 advanced GC patients linked with data on clinical response to ramucirumab or its combinations. Three genes showed differential expression in the tumors for responders versus nonresponders: CHRM3, LRFN1, and TEX15 Of them, CHRM3 was up-regulated in the responders. Using the bioinformatic platform Oncobox we simulated ramucirumab efficiency and compared output model results with actual tumor response data. An agreement was observed between predicted and real clinical outcomes (AUC ≥ 0.7). These results suggest that RNA sequencing may be used to personalize the prescription of ramucirumab for GC and indicate potential molecular mechanisms underlying ramucirumab resistance. The RNA sequencing profiles obtained here are fully compatible with the previously published Oncobox Atlas of Normal Tissue Expression (ANTE) data.

Standard machine learning approaches outperform deep representation learning on phenotype prediction from transcriptomics data

BACKGROUND

The ability to confidently predict health outcomes from gene expression would catalyze a revolution in molecular diagnostics. Yet, the goal of developing actionable, robust, and reproducible predictive signatures of phenotypes such as clinical outcome has not been attained in almost any disease area. Here, we report a comprehensive analysis spanning prediction tasks from ulcerative colitis, atopic dermatitis, diabetes, to many cancer subtypes for a total of 24 binary and multiclass prediction problems and 26 survival analysis tasks. We systematically investigate the influence of gene subsets, normalization methods and prediction algorithms. Crucially, we also explore the novel use of deep representation learning methods on large transcriptomics compendia, such as GTEx and TCGA, to boost the performance of state-of-the-art methods. The resources and findings in this work should serve as both an up-to-date reference on attainable performance, and as a benchmarking resource for further research.

RESULTS

Approaches that combine large numbers of genes outperformed single gene methods consistently and with a significant margin, but neither unsupervised nor semi-supervised representation learning techniques yielded consistent improvements in out-of-sample performance across datasets. Our findings suggest that using l2-regularized regression methods applied to centered log-ratio transformed transcript abundances provide the best predictive analyses overall.

CONCLUSIONS

Transcriptomics-based phenotype prediction benefits from proper normalization techniques and state-of-the-art regularized regression approaches. In our view, breakthrough performance is likely contingent on factors which are independent of normalization and general modeling techniques; these factors might include reduction of systematic errors in sequencing data, incorporation of other data types such as single-cell sequencing and proteomics, and improved use of prior knowledge.