From data to discovery: AI-guided analysis of disease-relevant molecules in spinal muscular atrophy (SMA).

Spinal Muscular Atrophy is caused by partial loss of survival of motoneuron (SMN) protein expression. The numerous interaction partners and mechanisms influenced by SMN loss result in a complex disease. Current treatments restore SMN protein levels to a certain extent, but do not cure all symptoms. The prolonged survival of patients creates an increasing need for a better understanding of SMA. Although many SMN-protein interactions, dysregulated pathways, and organ phenotypes are known, the connections among them remain largely unexplored. Monogenic diseases are ideal examples for the exploration of cause-and-effect relationships to create a network describing the disease-context. Machine learning tools can utilize such knowledge to analyze similarities between disease-relevant molecules and molecules not described in the disease so far. We used an artificial intelligence-based algorithm to predict new genes of interest. The transcriptional regulation of 8 out of 13 molecules selected from the predicted set were successfully validated in an SMA mouse model. This bioinformatic approach, using the given experimental knowledge for relevance predictions, enhances efficient targeted research in SMA and potentially in other disease settings.

Mining hidden knowledge: embedding models of cause-effect relationships curated from the biomedical literature.

Motivation: We explore the use of literature-curated signed causal gene expression and gene-function relationships to construct unsupervised embeddings of genes, biological functions and diseases. Our goal is to prioritize and predict activating and inhibiting functional associations of genes and to discover hidden relationships between functions. As an application, we are particularly interested in the automatic construction of networks that capture relevant biology in a given disease context. Results: We evaluated several unsupervised gene embedding models leveraging literature-curated signed causal gene expression findings. Using linear regression, we show that, based on these gene embeddings, gene-function relationships can be predicted with about 95% precision for the highest scoring genes. Function embedding vectors, derived from parameters of the linear regression model, allow inference of relationships between different functions or diseases. We show for several diseases that gene and function embeddings can be used to recover key drivers of pathogenesis, as well as underlying cellular and physiological processes. These results are presented as disease-centric networks of genes and functions. To illustrate the applicability of our approach to other machine learning tasks, we also computed embeddings for drug molecules, which were then tested using a simple neural network to predict drug-disease associations. Availability and implementation: Python implementations of the gene and function embedding algorithms operating on a subset of our literature-curated content as well as other code used for this paper are made available as part of the Supplementary data. Supplementary information: Supplementary data are available at Bioinformatics Advances online.

The Coronavirus Network Explorer: mining a large-scale knowledge graph for effects of SARS-CoV-2 on host cell function.

BACKGROUND: Leveraging previously identified viral interactions with human host proteins, we apply a machine learning-based approach to connect SARS-CoV-2 viral proteins to relevant host biological functions, diseases, and pathways in a large-scale knowledge graph derived from the biomedical literature. Our goal is to explore how SARS-CoV-2 could interfere with various host cell functions, and to identify drug targets amongst the host genes that could potentially be modulated against COVID-19 by repurposing existing drugs. The machine learning model employed here involves gene embeddings that leverage causal gene expression signatures curated from literature. In contrast to other network-based approaches for drug repurposing, our approach explicitly takes the direction of effects into account, distinguishing between activation and inhibition. RESULTS: We have constructed 70 networks connecting SARS-CoV-2 viral proteins to various biological functions, diseases, and pathways reflecting viral biology, clinical observations, and co-morbidities in the context of COVID-19. Results are presented in the form of interactive network visualizations through a web interface, the Coronavirus Network Explorer (CNE), that allows exploration of underlying experimental evidence. We find that existing drugs targeting genes in those networks are strongly enriched in the set of drugs that are already in clinical trials against COVID-19. CONCLUSIONS: The approach presented here can identify biologically plausible hypotheses for COVID-19 pathogenesis, explicitly connected to the immunological, virological and pathological observations seen in SARS-CoV-2 infected patients. The discovery of repurposable drugs is driven by prior knowledge of relevant functional endpoints that reflect known viral biology or clinical observations, therefore suggesting potential mechanisms of action. We believe that the CNE offers relevant insights that go beyond more conventional network approaches, and can be a valuable tool for drug repurposing. The CNE is available at https://digitalinsights.qiagen.com/coronavirus-network-explorer .

Pathways of Toxicity.

Despite wide-spread consensus on the need to transform toxicology and risk assessment in order to keep pace with technological and computational changes that have revolutionized the life sciences, there remains much work to be done to achieve the vision of toxicology based on a mechanistic foundation. To this end, a workshop was organized to explore one key aspect of this transformation - the development of Pathways of Toxicity as a key tool for hazard identification based on systems biology. Several issues were discussed in depth in the workshop: The first was the challenge of formally defining the concept of a Pathway of Toxicity (PoT), as distinct from, but complementary to, other toxicological pathway concepts such as mode of action (MoA). The workshop came up with a preliminary definition of PoT as "A molecular definition of cellular processes shown to mediate adverse outcomes of toxicants". It is further recognized that normal physiological pathways exist that maintain homeostasis and these, sufficiently perturbed, can become PoT. Second, the workshop sought to define the adequate public and commercial resources for PoT information, including data, visualization, analyses, tools, and use-cases, as well as the kinds of efforts that will be necessary to enable the creation of such a resource. Third, the workshop explored ways in which systems biology approaches could inform pathway annotation, and which resources are needed and available that can provide relevant PoT information to the diverse user communities.

Causal analysis approaches in Ingenuity Pathway Analysis.

MOTIVATION: Prior biological knowledge greatly facilitates the meaningful interpretation of gene-expression data. Causal networks constructed from individual relationships curated from the literature are particularly suited for this task, since they create mechanistic hypotheses that explain the expression changes observed in datasets. RESULTS: We present and discuss a suite of algorithms and tools for inferring and scoring regulator networks upstream of gene-expression data based on a large-scale causal network derived from the Ingenuity Knowledge Base. We extend the method to predict downstream effects on biological functions and diseases and demonstrate the validity of our approach by applying it to example datasets. AVAILABILITY: The causal analytics tools 'Upstream Regulator Analysis', 'Mechanistic Networks', 'Causal Network Analysis' and 'Downstream Effects Analysis' are implemented and available within Ingenuity Pathway Analysis (IPA, http://www.ingenuity.com). SUPPLEMENTARY INFORMATION: Supplementary material is available at Bioinformatics online.

Inhibitors of human indoleamine 2,3-dioxygenase identified with a target-based screen in yeast.

Indoleamine 2,3-dioxygenase (IDO) is a tryptophan degradation enzyme that is emerging as an important drug target. IDO is expressed by many human tumors to help them escape immune detection, and it has been implicated in depression and in the formation of senile nuclear cataracts. There is a need for potent and selective IDO inhibitors for use in research and as lead compounds for drug development. We show that expression of human IDO in a Saccharomyces cerevisiae tryptophan auxotroph restricts yeast growth in the presence of low tryptophan concentrations and that inhibition of IDO activity can restore growth. We use this assay to screen for IDO inhibitors in collections of pure chemicals and crude natural extracts. We identify NSC 401366 (imidodicarbonimidic diamide, N-methyl-N'-9-phenanthrenyl-, monohydrochloride) as a potent nonindolic IDO inhibitor (Ki=1.5 +/- 0.2 microM) that is competitive with respect to tryptophan. We also use this assay to identify the active compound caulerpin from a crude algal extract. The yeast growth restoration assay is simple and inexpensive. It combines desirable attributes of cell- and target-based screens and is an attractive tool for chemical biology and drug screening.

NSAID-induced acute phase response is due to increased intestinal permeability and characterized by early and consistent alterations in hepatic gene expression.

Toxicogenomics using a reference database can provide a better understanding and prediction of toxicity, largely by creating biomarkers that tie gene expression to actual pathology events. During the course of building a toxicogenomic database, an observation was made that a number of non-steroidal anti-inflammatory compounds (NSAIDs) at supra-pharmacologic doses induced an acute phase response (APR) and displayed hepatic gene expression patterns similar to that of intravenous lipopolysaccharide (LPS). Since NSAIDs are known to cause injury along the gastrointestinal tract, it has been suggested that NSAIDs increase intestinal permeability, allowing LPS and/or bacteria into the systemic circulation and stimulating an APR detectable in the liver. A short term study was subsequently conducted examining the effects of aspirin, indomethacin, ibuprofen, and rofecoxib to rats and a variety of endpoints were examined that included serum levels of inflammatory cytokines, histologic evaluation, and hepatic gene expression. Both indomethacin and ibuprofen injured the gastrointestinal tract, induced an APR, and increased serum levels of LPS, while rofecoxib and aspirin did not affect the GI tract or induce an APR. In treatments that eventually showed a systemic inflammatory response, hepatic expression of many inflammatory genes was noted as early as 6 hours after treatment well before alterations in traditional clinical pathology markers were detected. This finding led to the creation of a hepatic gene expression biomarker of APR that was effectively shown to be an early identifier of imminent inflammatory injury. In terms of the relative gastrointestinal safety and the NSAIDs studied, an important safety distinction can be made between the presumptive efficacious dose and the APR-inducing dose for indomethacin (1-2-fold), ibuprofen (5-fold), and rofecoxib (approximately 250-fold). Our data support the notion that NSAID-induced intestinal injury results in leakage of commensural bacteria and/or LPS into the circulation, provoking a systemic inflammatory response and that hepatic gene expression-based biomarkers can be used as early and sensitive biomarkers of APR onset.

Development of a large-scale chemogenomics database to improve drug candidate selection and to understand mechanisms of chemical toxicity and action.

Successful drug discovery requires accurate decision making in order to advance the best candidates from initial lead identification to final approval. Chemogenomics, the use of genomic tools in pharmacology and toxicology, offers a promising enhancement to traditional methods of target identification/validation, lead identification, efficacy evaluation, and toxicity assessment. To realize the value of chemogenomics information, a contextual database is needed to relate the physiological outcomes induced by diverse compounds to the gene expression patterns measured in the same animals. Massively parallel gene expression characterization coupled with traditional assessments of drug candidates provides additional, important mechanistic information, and therefore a means to increase the accuracy of critical decisions. A large-scale chemogenomics database developed from in vivo treated rats provides the context and supporting data to enhance and accelerate accurate interpretation of mechanisms of toxicity and pharmacology of chemicals and drugs. To date, approximately 600 different compounds, including more than 400 FDA approved drugs, 60 drugs approved in Europe and Japan, 25 withdrawn drugs, and 100 toxicants, have been profiled in up to 7 different tissues of rats (representing over 3200 different drug-dose-time-tissue combinations). Accomplishing this task required evaluating and improving a number of in vivo and microarray protocols, including over 80 rigorous quality control steps. The utility of pairing clinical pathology assessments with gene expression data is illustrated using three anti-neoplastic drugs: carmustine, methotrexate, and thioguanine, which had similar effects on the blood compartment, but diverse effects on hepatotoxicity. We will demonstrate that gene expression events monitored in the liver can be used to predict pathological events occurring in that tissue as well as in hematopoietic tissues.

Classification of a large microarray data set: algorithm comparison and analysis of drug signatures.

A large gene expression database has been produced that characterizes the gene expression and physiological effects of hundreds of approved and withdrawn drugs, toxicants, and biochemical standards in various organs of live rats. In order to derive useful biological knowledge from this large database, a variety of supervised classification algorithms were compared using a 597-microarray subset of the data. Our studies show that several types of linear classifiers based on Support Vector Machines (SVMs) and Logistic Regression can be used to derive readily interpretable drug signatures with high classification performance. Both methods can be tuned to produce classifiers of drug treatments in the form of short, weighted gene lists which upon analysis reveal that some of the signature genes have a positive contribution (act as "rewards" for the class-of-interest) while others have a negative contribution (act as "penalties") to the classification decision. The combination of reward and penalty genes enhances performance by keeping the number of false positive treatments low. The results of these algorithms are combined with feature selection techniques that further reduce the length of the drug signatures, an important step towards the development of useful diagnostic biomarkers and low-cost assays. Multiple signatures with no genes in common can be generated for the same classification end-point. Comparison of these gene lists identifies biological processes characteristic of a given class.

The dephosphorylated form of the anaphase-promoting complex protein Cdc27/Apc3 concentrates on kinetochores and chromosome arms in mitosis.

Cell cycle regulated protein ubiquitination and degradation within subcellular domains may be essential for the normal progression of mitosis. Cdc27 is a conserved component of an essential M-phase ubiquitin-protein ligase called the anaphase-promoting complex/cyclosome. We examined the subcellular distribution of Cdc27 in greater detail in mammalian cells and found Cdc27 concentrated at spindle poles and on spindle microtubules as previously described, but also found Cdc27 at kinetochores and along chromosome arms. This localization was not dependent on intact microtubules. While the great majority of Cdc27 protein in M phase cells is highly phosphorylated, only the dephosphorylated form of Cdc27 was found associated with isolated chromosomes. Kinases that also associate with isolated chromosomes catalyzed the in vitro phosphorylation of the chromosome-associated Cdc27. Microinjection of anti-Cdc27 antibody into cells causes arrest at metaphase. Microinjection of cells with anti-Mad2 antibody normally induces premature anaphase onset resulting in catastrophic nondisjunction of the chromosomes. However, coinjection of anti-Cdc27 antibody with anti-Mad2 antibody resulted in metaphase arrest. The association of dephosphorylated APC/C components with mitotic chromosomes suggests mechanisms by which the spindle checkpoint may regulate APC/C activity at mitosis.