Christos Andronis

Literature mining, ontologies and information visualization for drug repurposing

The immense growth of MEDLINE coupled with the realization that a vast amount of biomedical knowledge is recorded in free-text format, has led to the appearance of a large number of literature mining techniques aiming to extract biomedical terms and their inter-relations from the scientific literature. Ontologies have been extensively utilized in the biomedical domain either as controlled vocabularies or to provide the framework for mapping relations between concepts in biology and medicine. Literature-based approaches and ontologies have been used in the past for the purpose of hypothesis generation in connection with drug discovery. Here, we review the application of literature mining and ontology modeling and traversal to the area of drug repurposing (DR). In recent years, DR has emerged as a noteworthy alternative to the traditional drug development process, in response to the decreased productivity of the biopharmaceutical industry. Thus, systematic approaches to DR have been developed, involving a variety of in silico, genomic and high-throughput screening technologies. Attempts to integrate literature mining with other types of data arising from the use of these technologies as well as visualization tools assisting in the discovery of novel associations between existing drugs and new indications will also be presented.

Mining of relations between proteins over biomedical scientific literature using a deep-linguistic approach

OBJECTIVE The amount of new discoveries (as published in the scientific literature) in the biomedical area is growing at an exponential rate. This growth makes it very difficult to filter the most relevant results, and thus the extraction of the core information becomes very expensive. Therefore, there is a growing interest in text processing approaches that can deliver selected information from scientific publications, which can limit the amount of human intervention normally needed to gather those results. MATERIALS AND METHODS This paper presents and evaluates an approach aimed at automating the process of extracting functional relations (e.g. interactions between genes and proteins) from scientific literature in the biomedical domain. The approach, using a novel dependency-based parser, is based on a complete syntactic analysis of the corpus. RESULTS We have implemented a state-of-the-art text mining system for biomedical literature, based on a deep-linguistic, full-parsing approach. The results are validated on two different corpora: the manually annotated genomics information access (GENIA) corpus and the automatically annotated arabidopsis thaliana circadian rhythms (ATCR) corpus. CONCLUSION We show how a deep-linguistic approach (contrary to common belief) can be used in a real world text mining application, offering high-precision relation extraction, while at the same time retaining a sufficient recall.

Drug repurposing and adverse event prediction using high-throughput literature analysis.

Drug repurposing is the process of using existing drugs in indications other than the ones they were originally designed for. It is an area of significant recent activity due to the mounting costs of traditional drug development and scarcity of new chemical entities brought to the market by bio‐pharmaceutical companies. By selecting drugs that already satisfy basic toxicity, ADME and related criteria, drug repurposing promises to deliver significant value at reduced cost and in dramatically shorter time frames than is normally the case for the drug development process. The same process that results in drug repurposing can also be used for the prediction of adverse events of known or novel drugs. The analytics method is based on the description of the mechanism of action of a drug, which is then compared to the molecular mechanisms underlying all known adverse events. This review will focus on those approaches to drug repurposing and adverse event prediction that are based on the biomedical literature. Such approaches typically begin with an analysis of the literature and aim to reveal indirect relationships among seemingly unconnected biomedical entities such as genes, signaling pathways, physiological processes, and diseases. Networks of associations of these entities allow the uncovering of the molecular mechanisms underlying a disease, better understanding of the biological effects of a drug and the evaluation of its benefit/risk profile. In silico results can be tested in relevant cellular and animal models and, eventually, in clinical trials. WIREs Syst Biol Med 2011 3 323–334 DOI: 10.1002/wsbm.147

From depression to neurodegeneration and heart failure: re-examining the potential of MAO inhibitors

Initially introduced in the 1950s for treating depression, monoamine oxidase (MAO) inhibitors were gradually abandoned, mainly owing to their potential for drug–drug and drug–food interactions, the most widely known being with tyramine-containing food (the ‘cheese’ effect). Since then, more selective MAO-A or MAO-B inhibitors have been developed with substantially reduced risks, and have been approved for the treatment of depression and Parkinson’s disease, respectively. Recent research suggests that some of these drugs also have neuroprotective properties, while preclinical evidence expands the spectrum of potential indications to heart failure, renal diseases and multiple sclerosis. In this article, the authors review the relevance of MAO isoforms to disease, and they also outline current research and development efforts in this class of drugs, including newer multipotent compounds.

Relation mining over a corpus of scientific literature

The amount of new discoveries (as published in the scientific literature) in the area of Molecular Biology is currently growing at an exponential rate. This growth makes it very difficult to filter the most relevant results, and the extraction of the core information, for inclusion in one of the knowledge resources being maintained by the research community, becomes very expensive. Therefore, there is a growing interest in text processing approaches that can deliver selected information from scientific publications, which can limit the amount of human intervention normally needed to gather those results. This paper presents and evaluates an approach aimed at automating the process of extracting semantic relations (e.g. interactions between genes and proteins) from scientific literature in the domain of Molecular Biology. The approach, using a novel dependency-based parser, is based on a complete syntactic analysis of the corpus.

Discordant effects of rasagiline doses in Parkinson disease

we read with interest the news & views article by sampaio et al. (Parkinson disease: aDaGiO trial hints that rasagiline slows disease progression. Nat. Rev. Neurol. 6, 126–128; 2010).1 the authors conclude that although the net benefit that rasagiline is able to deliver is what is really relevant for an individual patient’s wellbeing, understanding the diseasemodifying effects of the drug has much to offer to the scientific community and to society at large. we agree with their view and we believe that elucidating the discordant neuroprotective effects of the 1 mg/day and 2 mg/day doses of rasagiline remains important, in view of the continuing scientific debate on the topic.2 the authors of the aDaGiO study found the fact that the higher dose of rasagiline did not confer a stronger neuroprotective effect than the lower dose difficult to explain and proposed that a marked effect of the higher dose on symptoms might have masked a benefit associated with early-start treatment in patients with very mild disease.2,3 However, rasagiline has several mechanisms of action in addition to monoamine oxidase type B inhibition, which makes alternative explanations possible. in particular, rasagiline has been reported to induce the anti-apoptotic protein Bcl-2 at 10 nM to 10 μM and 10–100 pM but not outside these ranges, and to upregulate glial cell line-derived neurotrophic factor (GDnF) at 100 nM.2,4 Considering that the maximum plasma concentrations after repeated dosing with 1 mg/day or 2 mg/day are 8.5 ± 2.2 ng/ml (31.7 ± 8.2nM) and 14.9 ± 10.5ng/ml (=55.7 ± 39.2nM), respectively,5 the higher dose would be expected to exert a more potent neuroprotective effect than the lower one, contrary to the observed clinical results. therefore, rasagiline neuroprotection is unlikely to be mediated solely by upregulation of GDnF or Bcl-2. recently, rasagiline has been shown to inhibit glyceraldehyde-3-phosphate dehydrogenase (GaPDH), an enzyme that catalyzes the sixth step of glycolysis and participates in the initiation of apoptosis.6 under oxidative stress, GaPDH is inactivated, causing a switch of glucose metabolism to the pentose phosphate pathway. this increases the production of naDPH, which is required for the recycling of glutathione. inhibition of GaPDH, therefore, enhances the antioxidant defenses of dopaminergic cells, and protects them from apoptosis. However, by inhibiting glycolysis, the drug also deprives cells of much-needed energy. we propose that the balance between enhancement of the anti-apoptotic/ antioxidant effects and inhibition of the glycolytic actions of GaPDH by rasagiline may contribute towards determining its overall effects. stronger inhibition of glycolysis at 2 mg/day might outweigh other benefits, causing reduced clinical efficacy. we believe that additional studies are warranted to examine this hypothesis, which might help elucidate the puzzling effects of rasagiline.

Binary Relation Extraction from Biomedical Literature using Dependency Trees and SVMs

The goal of Biomedical relation extraction is to uncover high-quality relations from life science literature with diverse applications in the fields of Biology and Medicine. In the last decade, several methods can be found in published literature ranging from binary to complex relation extraction. In this work, we present a binary relation extraction system that relies on sentence level dependency features. We use a novel approach to map dependency tree based rules to feature vectors that can be used to train a classifier. We build a SVM classifier using these feature vectors and our experimental results show that it outperforms simple co-occurrence and rule-based systems. Through our experiments, using two ‘real-world’ examples, we quantify the positive impact of improved relation extraction on Literature Based Discovery.

Identification of echocardiographic indices for the early detection of left-ventricular systolic dysfunction in beta-thalassaemia via Self-Organizing Maps: A data-exploration study

Congestive heart failure (CHF) remains the primary cause of death in patients suffering from beta-thalassaemia major. Its early detection allows the prompt initiation of aggressive chelation therapy, when the condition can still be reversed. We aimed at identifying echocardiographic indices for the early detection of left ventricular (LV) systolic dysfunction, the physiological abnormality underlying CHF, in these patients. We used Self-Organizing Maps (SOMs)—an artificial neural network—for identifying novel correlations within our Electronic Healthcare Record (EHCR) database on beta-thalassaemia. We sought echocardiographic parameters that are correlated to future deterioration of the LV ejection fraction and therefore constitute early signs of LV systolic dysfunction. At the same time, we evaluated SOMs as tools for exploring clinical datasets and make recommendations on the setup of the SOM algorithm that is appropriate for such tasks. We found that high values of the LV end-systolic diameter index and of the E/A ratio are early indications of LV systolic dysfunction. From a technical point of view, zero-mean unit-variance normalization of the input data, a large initial neighbourhood radius and a rectangular SOM grid produced optimal maps for the purpose of detecting clinical correlations. We have successfully used SOMs for exploring a clinical dataset and for creating novel medical hypotheses. A clinical study has been launched to confirm these hypotheses, and initial results are encouraging.