Early Detection of Macular Atrophy Automated Through 2D and 3D Unet Deep Learning.

Macular atrophy (MA) is an irreversible endpoint of age-related macular degeneration (AMD), which is the leading cause of blindness in the world. Early detection is therefore an unmet need. We have developed a novel automated method to identify MA in patients undergoing follow-up with optical coherence tomography (OCT) for AMD based on the combination of 2D and 3D Unet architecture.

Genomic-driven nutritional interventions for radiotherapy-resistant rectal cancer patient.

Radiotherapy response of rectal cancer patients is dependent on a myriad of molecular mechanisms including response to stress, cell death, and cell metabolism. Modulation of lipid metabolism emerges as a unique strategy to improve radiotherapy outcomes due to its accessibility by bioactive molecules within foods. Even though a few radioresponse modulators have been identified using experimental techniques, trying to experimentally identify all potential modulators is intractable.

Alzheimer's disease: using gene/protein network machine learning for molecule discovery in olive oil.

Alzheimer's disease (AD) poses a profound human, social, and economic burden. Previous studies suggest that extra virgin olive oil (EVOO) may be helpful in preventing cognitive decline. Here, we present a network machine learning method for identifying bioactive phytochemicals in EVOO with the highest potential to impact the protein network linked to the development and progression of the AD.

Virus-induced Volatile Organic Compounds Are Detectable in Exhaled Breath during Pulmonary Infection.

Chronic obstructive pulmonary disease (COPD) is a condition punctuated by acute exacerbations commonly triggered by viral and/or bacterial infection. Early identification of exacerbation triggers is important to guide appropriate therapy, but currently available tests are slow and imprecise. Volatile organic compounds (VOCs) can be detected in exhaled breath and have the potential to be rapid tissue-specific biomarkers of infection etiology.

Predicting anticancer hyperfoods with graph convolutional networks.

Background: Recent efforts in the field of nutritional science have allowed the discovery of disease-beating molecules within foods based on the commonality of bioactive food molecules to FDA-approved drugs. The pioneering work in this field used an unsupervised network propagation algorithm to learn the systemic-wide effect on the human interactome of 1962 FDA-approved drugs and a supervised algorithm to predict anticancer therapeutics using the learned representations. Then, a set of bioactive molecules within foods was fed into the model, which predicted molecules with cancer-beating potential.

Network machine learning maps phytochemically rich "Hyperfoods" to fight COVID-19.

In this paper, we introduce a network machine learning method to identify potential bioactive anti-COVID-19 molecules in foods based on their capacity to target the SARS-CoV-2-host gene-gene (protein-protein) interactome. Our analyses were performed using a supercomputing DreamLab App platform, harnessing the idle computational power of thousands of smartphones. Machine learning models were initially calibrated by demonstrating that the proposed method can predict anti-COVID-19 candidates among experimental and clinically approved drugs (5658 in total) targeting COVID-19 interactomics with the balanced classification accuracy of 80-85% in 5-fold cross-validated settings.

Auto-deconvolution and molecular networking of gas chromatography-mass spectrometry data.

We engineered a machine learning approach, MSHub, to enable auto-deconvolution of gas chromatography-mass spectrometry (GC-MS) data. We then designed workflows to enable the community to store, process, share, annotate, compare and perform molecular networking of GC-MS data within the Global Natural Product Social (GNPS) Molecular Networking analysis platform. MSHub/GNPS performs auto-deconvolution of compound fragmentation patterns via unsupervised non-negative matrix factorization and quantifies the reproducibility of fragmentation patterns across samples.

HyperFoods: Machine intelligent mapping of cancer-beating molecules in foods.

Recent data indicate that up-to 30-40% of cancers can be prevented by dietary and lifestyle measures alone. Herein, we introduce a unique network-based machine learning platform to identify putative food-based cancer-beating molecules. These have been identified through their molecular biological network commonality with clinically approved anti-cancer therapies.

Trapping of the transport-segment DNA by the ATPase domains of a type II topoisomerase.

Type II topoisomerases alter DNA topology to control DNA supercoiling and chromosome segregation and are targets of clinically important anti-infective and anticancer therapeutics. They act as ATP-operated clamps to trap a DNA helix and transport it through a transient break in a second DNA. Here, we present the first X-ray crystal structure solved at 2.

Exploiting and assessing multi-source data for supervised biomedical named entity recognition.

Motivation: Recognition of biomedical entities from scientific text is a critical component of natural language processing and automated information extraction platforms. Modern named entity recognition approaches rely heavily on supervised machine learning techniques, which are critically dependent on annotated training corpora. These approaches have been shown to perform well when trained and tested on the same source.

BASIS: High-performance bioinformatics platform for processing of large-scale mass spectrometry imaging data in chemically augmented histology.

Mass Spectrometry Imaging (MSI) holds significant promise in augmenting digital histopathologic analysis by generating highly robust big data about the metabolic, lipidomic and proteomic molecular content of the samples. In the process, a vast quantity of unrefined data, that can amount to several hundred gigabytes per tissue section, is produced. Managing, analysing and interpreting this data is a significant challenge and represents a major barrier to the translational application of MSI.

ChemDistiller: an engine for metabolite annotation in mass spectrometry.

Motivation: High-resolution mass spectrometry permits simultaneous detection of thousands of different metabolites in biological samples; however, their automated annotation still presents a challenge due to the limited number of tailored computational solutions freely available to the scientific community.

Results: Here, we introduce ChemDistiller, a customizable engine that combines automated large-scale annotation of metabolites using tandem MS data with a compiled database containing tens of millions of compounds with pre-calculated 'fingerprints' and fragmentation patterns. Our tests using publicly and commercially available tandem MS spectra for reference compounds show retrievals rates comparable to or exceeding the ones obtainable by the current state-of-the-art solutions in the field while offering higher throughput, scalability and processing speed.

Translational utility of a hierarchical classification strategy in biomolecular data analytics.

Hierarchical classification (HC) stratifies and classifies data from broad classes into more specific classes. Unlike commonly used data classification strategies, this enables the probabilistic prediction of unknown classes at different levels, minimizing the burden of incomplete databases. Despite these advantages, its translational application in biomedical sciences has been limited.

Exploring the active site of the Streptococcus pneumoniae topoisomerase IV-DNA cleavage complex with novel 7,8-bridged fluoroquinolones.

As part of a programme of synthesizing and investigating the biological properties of new fluoroquinolone antibacterials and their targeting of topoisomerase IV from Streptococcus pneumoniae, we have solved the X-ray structure of the complexes of two new 7,8-bridged fluoroquinolones (with restricted C7 group rotation favouring tight binding) in complex with the topoisomerase IV from S. pneumoniae and an 18-base-pair DNA binding site-the E-site-found by our DNA mapping studies to bind drug strongly in the presence of topoisomerase IV (Leo et al. 2005 J.

Structure of a quinolone-stabilized cleavage complex of topoisomerase IV from Klebsiella pneumoniae and comparison with a related Streptococcus pneumoniae complex.

Klebsiella pneumoniae is a Gram-negative bacterium that is responsible for a range of common infections, including pulmonary pneumonia, bloodstream infections and meningitis. Certain strains of Klebsiella have become highly resistant to antibiotics. Despite the vast amount of research carried out on this class of bacteria, the molecular structure of its topoisomerase IV, a type II topoisomerase essential for catalysing chromosomal segregation, had remained unknown.

Interceptor effect of C60 fullerene on the in vitro action of aromatic drug molecules.

C60 fullerenes are spherical molecules composed purely of carbon atoms. They inspire a particularly strong scientific interest because of their specific physico-chemical properties and potential medical and nanotechnological applications. In this work we are focusing on studying the influence of the pristine C60 fullerene on biological activity of some aromatic drug molecules in human buccal epithelial cells.

Structure of an 'open' clamp type II topoisomerase-DNA complex provides a mechanism for DNA capture and transport.

Type II topoisomerases regulate DNA supercoiling and chromosome segregation. They act as ATP-operated clamps that capture a DNA duplex and pass it through a transient DNA break in a second DNA segment via the sequential opening and closure of ATPase-, G-DNA- and C-gates. Here, we present the first 'open clamp' structures of a 3-gate topoisomerase II-DNA complex, the seminal complex engaged in DNA recognition and capture.

Structural basis of gate-DNA breakage and resealing by type II topoisomerases.

Type II DNA topoisomerases are ubiquitous enzymes with essential functions in DNA replication, recombination and transcription. They change DNA topology by forming a transient covalent cleavage complex with a gate-DNA duplex that allows transport of a second duplex though the gate. Despite its biological importance and targeting by anticancer and antibacterial drugs, cleavage complex formation and reversal is not understood for any type II enzyme.

Structural insight into the quinolone-DNA cleavage complex of type IIA topoisomerases.

Type II topoisomerases alter DNA topology by forming a covalent DNA-cleavage complex that allows DNA transport through a double-stranded DNA break. We present the structures of cleavage complexes formed by the Streptococcus pneumoniae ParC breakage-reunion and ParE TOPRIM domains of topoisomerase IV stabilized by moxifloxacin and clinafloxacin, two antipneumococcal fluoroquinolones. These structures reveal two drug molecules intercalated at the highly bent DNA gate and help explain antibacterial quinolone action and resistance.

The difficult case of crystallization and structure solution for the ParC55 breakage-reunion domain of topoisomerase IV from Streptococcus pneumoniae.

Background: Streptococcus pneumoniae is the major cause of community-acquired pneumonia and is also associated with bronchitis, meningitis, otitis and sinusitis. The emergence and increasing prevalence of resistance to penicillin and other antibiotics has led to interest in other anti-pneumonococcal drugs such as quinolones that target the enzymes DNA gyrase and topoisomerase IV. During crystallization and in the avenues to finding a method to determine phases for the structure of the ParC55 breakage-reunion domain of topoisomerase IV from Streptococcus pneumoniae, obstacles were faced at each stage of the process.

Breakage-reunion domain of Streptococcus pneumoniae topoisomerase IV: crystal structure of a gram-positive quinolone target.

The 2.7 A crystal structure of the 55-kDa N-terminal breakage-reunion domain of topoisomerase (topo) IV subunit A (ParC) from Streptococcus pneumoniae, the first for the quinolone targets from a gram-positive bacterium, has been solved and reveals a 'closed' dimer similar in fold to Escherichia coli DNA gyrase subunit A (GyrA), but distinct from the 'open' gate structure of Escherichia coli ParC. Unlike GyrA whose DNA binding groove is largely positively charged, the DNA binding site of ParC exhibits a distinct pattern of alternating positively and negatively charged regions coincident with the predicted positions of the grooves and phosphate backbone of DNA.