Big data and machine learning for materials science.

Herein, we review aspects of leading-edge research and innovation in materials science that exploit big data and machine learning (ML), two computer science concepts that combine to yield computational intelligence. ML can accelerate the solution of intricate chemical problems and even solve problems that otherwise would not be tractable. However, the potential benefits of ML come at the cost of big data production; that is, the algorithms demand large volumes of data of various natures and from different sources, from material properties to sensor data. In the survey, we propose a roadmap for future developments with emphasis on computer-aided discovery of new materials and analysis of chemical sensing compounds, both prominent research fields for ML in the context of materials science. In addition to providing an overview of recent advances, we elaborate upon the conceptual and practical limitations of big data and ML applied to materials science, outlining processes, discussing pitfalls, and reviewing cases of success and failure.

Skeletal Optimization of Cytotoxic Lipidic Dialkynylcarbinols.

In line with a recent study of the pharmacological potential of bioinspired synthetic acetylenic lipids, after identification of the terminal dialkynylcarbinol (DAC) and butadiynyl alkynylcarbinol (BAC) moieties as functional antitumor pharmacophoric units, this work specifically addresses the issue of carbon backbone length. A systematic variation of the aliphatic chain length was thus carried out in both the DAC and BAC series. The critical impact of the length of the lipidic skeleton was first confirmed in the racemic series, with the highest cytotoxic activity observed for C17 to C18 backbones. Enantiomerically enriched samples were prepared by asymmetric synthesis of the optimal C18 DAC and C17 BAC derivatives. Samples with upgraded enantiomeric purity were alternatively produced by enzymatic kinetic resolution. Eutomers possessing the S configuration displayed cytotoxicity IC50 values as low as 15 nm against HCT116 cancer cells, the highest level of activity reached to date in this series.

Annonalide and derivatives: Semisynthesis, cytotoxic activities and studies on interaction of annonalide with DNA.

The cytotoxic activity of the pimarane diterpene annonalide (1) and nine of its semisynthetic derivatives (2-10) was investigated against the human tumor cell lines HL-60 (leukemia), PC-3 (prostate adenocarcinoma), HepG2 (hepatocellular carcinoma), SF-295 (glioblastoma) and HCT-116 (colon cancer), and normal mouse fibroblast (L929) cells. The preparation of 2-10 involved derivatization of the side chain of 1 at C-13. Except for 2, all derivatives are being reported for the first time. Most of the tested compounds presented IC50s below 4.0 µM, being considered potential antitumor agents. The structures of all new compounds were elucidated by spectroscopic analyses including 2D NMR and HRMS. Additionally, the interaction of annonalide (1) with ctDNA was evaluated using spectroscopic techniques, and the formation of a supramolecular complex with the macromolecule was confirmed. Competition assays with fluorescent probes (Hoechst and ethidium bromide) and theoretical studies confirmed that 1 interacts preferentially via DNA intercalation with stoichiometric ratio of 1:1 (1:ctDNA). The ΔG value was calculated as -28.24 kJ mol-1, and indicated that the interaction process occurs spontaneously. Docking studies revealed that van der Walls is the most important interaction in 1-DNA and EB-DNA complexes, and that both ligands (1 and EB) interact with the same DNA residues (DA6, DA17 and DT19).

Modulation of polyketide biosynthetic pathway of the endophytic fungus, Anteaglonium sp. FL0768, by copper (II) and anacardic acid.

In an attempt to explore the biosynthetic potential of endosymbiotic fungi, the secondary metabolite profiles of the endophytic fungus, Anteaglonium sp. FL0768, cultured under a variety of conditions were investigated. In potato dextrose broth (PDB) medium, Anteaglonium sp. FL0768 produced the heptaketides, herbaridine A (1), herbarin (2), 1-hydroxydehydroherbarin (3), scorpinone (4), and the methylated hexaketide 9S,11R-(+)-ascosalitoxin (5). Incorporation of commonly used epigenetic modifiers, 5-azacytidine and suberoylanilide hydroxamic acid, into the PDB culture medium of this fungus had no effect on its secondary metabolite profile. However, the histone acetyl transferase inhibitor, anacardic acid, slightly affected the metabolite profile affording scorpinone (4) as the major metabolite together with 1-hydroxydehydroherbarin (3) and a different methylated hexaketide, ascochitine (6). Intriguingly, incorporaion of Cu2+ into the PDB medium enhanced production of metabolites and drastically affected the biosynthetic pathway resulting in the production of pentaketide dimers, palmarumycin CE4 (7), palmarumycin CP4 (8), and palmarumycin CP1 (9), in addition to ascochitine (6). The structure of the new metabolite 7 was established with the help of spectroscopic data and by MnO2 oxidation to the known pentaketide dimer, palmarumycin CP3 (10). Biosynthetic pathways to some metabolites in Anteaglonium sp. FL0768 are presented and possible effects of AA and Cu2+ on these pathways are discussed.

Secondary metabolite production by the basidiomycete, Lentinus strigellus, under different culture conditions.

The basidiomycete Lentinus strigellus was cultivated in three different culture media and the secondary metabolites produced under different culture conditions were isolated and identified. When cultivated in a liquid medium with peptone, L. strigellus afforded the benzopyrans, 2,2-dimethyl-6-methoxychroman-4-one, 4-hydroxy-2,2-dimethyl-6-methoxychromane and (3R,4S)-3,4-dihydroxy-2,2-dimethyl-6-methoxychromane. The indole alkaloid echinuline and the anthraquinone fiscione, both unprecedented for the genus Lentinus, were isolated from the mycelium of the fungus. When cultured in Czapek medium enriched with potato broth, the fungus afforded the same benzopyrans except (3S,4S)-3,4-dihydroxy-2,2-dimethyl-6-methoxychromane. Panepoxydone and isopanepoxydone were also isolated when the microorganism was grown in Czapek medium.

Using multidimensional projection techniques for reaching a high distinguishing ability in biosensing.

Recent advances in the control of molecular engineering architectures have allowed unprecedented ability of molecular recognition in biosensing, with a promising impact for clinical diagnosis and environment control. The availability of large amounts of data from electrical, optical, or electrochemical measurements requires, however, sophisticated data treatment in order to optimize sensing performance. In this study, we show how an information visualization system based on projections, referred to as Projection Explorer (PEx), can be used to achieve high performance for biosensors made with nanostructured films containing immobilized antigens. As a proof of concept, various visualizations were obtained with impedance spectroscopy data from an array of sensors whose electrical response could be specific toward a given antibody (analyte) owing to molecular recognition processes. In addition to discussing the distinct methods for projection and normalization of the data, we demonstrate that an excellent distinction can be made between real samples tested positive for Chagas disease and Leishmaniasis, which could not be achieved with conventional statistical methods. Such high performance probably arose from the possibility of treating the data in the whole frequency range. Through a systematic analysis, it was inferred that Sammon's mapping with standardization to normalize the data gives the best results, where distinction could be made of blood serum samples containing 10(-7) mg/mL of the antibody. The method inherent in PEx and the procedures for analyzing the impedance data are entirely generic and can be extended to optimize any type of sensor or biosensor.

Biosensors for efficient diagnosis of leishmaniasis: innovations in bioanalytics for a neglected disease.

The need for reliable, fast diagnostics is closely linked to the need for safe, effective treatment of the so-called "neglected" diseases. The list of diseases with no field-adapted diagnostic tools includes leishmaniasis, shigella, typhoid, and bacterial meningitis. Leishmaniasis, in particular, is a parasitic disease caused by Leishmania spp. transmitted by infected phlebotomine sandfly, which remains a public health concern in developing countries with ca. 12 million people infected and 350 million at risk of infection. Despite several attempts, methods for diagnosis are still noneffective, especially with regard to specificity due to false positives with Chagas' disease caused by Trypanosoma cruzi . Accepted golden standards for detecting leishmaniasis involve isolation of parasites either microscopically, or by culture, and in both methods specimens are obtained by invasive means. Here, we show that efficient distinction between cutaneous leishmaniasis and Chagas' disease can be obtained with a low-cost biosensor system made with nanostructured films containing specific Leishmania amazonensis and T. cruzi antigens and employing impedance spectroscopy as the detection method. This unprecedented selectivity was afforded by antigen-antibody molecular recognition processes inherent in the detection with the immobilized antigens, and by statistically correlating the electrical impedance data, which allowed distinction between real samples that tested positive for Chagas' disease and leishmaniasis. Distinction could be made of blood serum samples containing 10(-5) mg/mL of the antibody solution in a few minutes. The methods used here are generic and can be extended to any type of biosensor, which is important for an effective diagnosis of many other diseases.

Use of information visualization methods eliminating cross talk in multiple sensing units investigated for a light-addressable potentiometric sensor.

The integration of nanostructured films containing biomolecules and silicon-based technologies is a promising direction for reaching miniaturized biosensors that exhibit high sensitivity and selectivity. A challenge, however, is to avoid cross talk among sensing units in an array with multiple sensors located on a small area. In this letter, we describe an array of 16 sensing units of a light-addressable potentiometric sensor (LAPS), which was made with layer-by-layer (LbL) films of a poly(amidomine) dendrimer (PAMAM) and single-walled carbon nanotubes (SWNTs), coated with a layer of the enzyme penicillinase. A visual inspection of the data from constant-current measurements with liquid samples containing distinct concentrations of penicillin, glucose, or a buffer indicated a possible cross talk between units that contained penicillinase and those that did not. With the use of multidimensional data projection techniques, normally employed in information visualization methods, we managed to distinguish the results from the modified LAPS, even in cases where the units were adjacent to each other. Furthermore, the plots generated with the interactive document map (IDMAP) projection technique enabled the distinction of the different concentrations of penicillin, from 5 mmol L(-1) down to 0.5 mmol L(-1). Data visualization also confirmed the enhanced performance of the sensing units containing carbon nanotubes, consistent with the analysis of results for LAPS sensors. The use of visual analytics, as with projection methods, may be essential to handle a large amount of data generated in multiple sensor arrays to achieve high performance in miniaturized systems.

Antioxidant and larvicidal activities of Tephrosia egregia Sandw against Aedes aegypti.

The antioxidant and larvicidal activities of Tephrosia egregia extracts and its major component, dehydrorotenone, were studied. High antioxidant activity was found for dehydrorotenone and methanol and ethyl acetate extracts from roots and stems, respectively. Among the tested extracts, the hexane extract from stems showed potent larvicidal activity (LC50 12.88 +/- 0.64) while low activity was found for dehydrorotenone.