Missing value imputation strategies for metabolomics data

The origin of missing values can be caused by different reasons and depending on these origins missing values should be considered differently and dealt with in different ways. In this research, four methods of imputation have been compared with respect to revealing their effects on the normality and variance of data, on statistical significance and on the approximation of a suitable threshold to accept missing data as truly missing. Additionally, the effects of different strategies for controlling familywise error rate or false discovery and how they work with the different strategies for missing value imputation have been evaluated. Missing values were found to affect normality and variance of data and k‐means nearest neighbour imputation was the best method tested for restoring this. Bonferroni correction was the best method for maximizing true positives and minimizing false positives and it was observed that as low as 40% missing data could be truly missing. The range between 40 and 70% missing values was defined as a “gray area” and therefore a strategy has been proposed that provides a balance between the optimal imputation strategy that was k‐means nearest neighbor and the best approximation of positioning real zeros.     

Expanding Lipidome Coverage Using LC-MS/MS Data-Dependent Acquisition with Automated Exclusion List Generation

Untargeted omics analyses aim to comprehensively characterize biomolecules within a biological system. Changes in the presence or quantity of these biomolecules can indicate important biological perturbations, such as those caused by disease. With current technological advancements, the entire genome can now be sequenced; however, in the burgeoning fields of lipidomics, only a subset of lipids can be identified. The recent emergence of high resolution tandem mass spectrometry (HR-MS/MS), in combination with ultra-high performance liquid chromatography, has resulted in an increased coverage of the lipidome. Nevertheless, identifications from MS/MS are generally limited by the number of precursors that can be selected for fragmentation during chromatographic elution. Therefore, we developed the software IE-Omics to automate iterative exclusion (IE), where selected precursors using data-dependent topN analyses are excluded in sequential injections. In each sequential injection, unique precursors are fragmented until HR-MS/MS spectra of all ions above a user-defined intensity threshold are acquired. IE-Omics was applied to lipidomic analyses in Red Cross plasma and substantia nigra tissue. Coverage of the lipidome was drastically improved using IE. When applying IE-Omics to Red Cross plasma and substantia nigra lipid extracts in positive ion mode, 69% and 40% more molecular identifications were obtained, respectively. In addition, applying IE-Omics to a lipidomics workflow increased the coverage of trace species, including odd-chained and short-chained diacylglycerides and oxidized lipid species. By increasing the coverage of the lipidome, applying IE to a lipidomics workflow increases the probability of finding biomarkers and provides additional information for determining etiology of disease.

Open image in new window

Graphical Abstract ?

     

Hybrid fluorescent nanoparticles from quantum dots coupled to cellulose nanocrystals

Carboxylated cellulose nanocrystals (CNCs) were decorated with CdSe/ZnS quantum dots (QDs) using a carbodiimide chemistry coupling approach. The one-step covalent modification was supported by nanoscale imaging, which showed QDs clustered on and around the CNCs after coupling. The QD–CNC hybrid nanoparticles remained colloidally stable in aqueous suspension and were fluorescent, exhibiting the broad excitation and narrow emission profile characteristic of the QDs. QD–CNCs in nanocomposite films imparted strong fluorescence within CNC-compatible matrices at relatively low loadings (0.15 nmol QDs/g of dry film), without altering the overall physical properties or self-assembly of the CNCs. The hybrid QD–CNCs may find applications in nanoparticle tracking, bio-imaging, optical/sensing devices, and anti-counterfeit technologies.     

Circumference and Pathwidth of Highly Connected Graphs

Birmele [J Graph Theory 2003] proved that every graph with circumference t has treewidth at most . Under the additional assumption of 2‐connectivity, such graphs have bounded pathwidth, which is a qualitatively stronger conclusion. Birmele's theorem was extended by Birmele et al. [Combinatorica 2007] who showed that every graph without k disjoint cycles of length at least t has treewidth . Our main result states that, under the additional assumption of ‐connectivity, such graphs have bounded pathwidth. In fact, they have pathwidth . Moreover, examples show that ‐connectivity is required for bounded pathwidth to hold. These results suggest the following general question: for which values of k and graphs H does every k‐connected H‐minor‐free graph have bounded pathwidth? We discuss this question and provide a few observations.     

Wrapping of nanoparticles by membranes

How nanoparticles interact with biomembranes is central for understanding their bioactivity. Biomembranes wrap around nanoparticles if the adhesive interaction between the nanoparticles and membranes is sufficiently strong to compensate for the cost of membrane bending. In this article, we review recent results from theory and simulations that provide new insights on the interplay of bending and adhesion energies during the wrapping of nanoparticles by membranes. These results indicate that the interplay of bending and adhesion during wrapping is strongly affected by the interaction range of the particle–membrane adhesion potential, by the shape of the nanoparticles, and by shape changes of membrane vesicles during wrapping. The interaction range of the particle–membrane adhesion potential is crucial both for the wrapping process of single nanoparticles and the cooperative wrapping of nanoparticles by membrane tubules.     

High angular-resolution automated visible-wavelength scanning angle Raman microscopy

A scanning angle (SA) Raman microscope with 532-nm excitation is reported for probing chemical content perpendicular to a sample interface. The instrument is fully automated to collect Raman spectra across a range of incident angles from 20.50 to 79.50° with an angular spread of 0.4 ± 0.2° and an angular uncertainty of 0.09°. Instrumental controls drive a rotational stage with a fixed axis of rotation relative to a prism-based sample interface mounted on an inverted microscope stage. Three benefits of SA Raman microscopy using visible wavelengths, compared to near infrared wavelengths are: (i) better surface sensitivity; (ii) increased signal due to the frequency to the fourth power dependence of the Raman signal, and the possibility for resonant enhancement; (iii) the need to scan a reduced angular range to shorten data collection times. These benefits were demonstrated with SA Raman measurements of thin polymer films of polystyrene or a diblock copolymer of polystyrene and poly(3-hexylthiophene-2,5-diyl). Thin film spectra were collected with a signal-to-noise ratio of 30 using a 0.25 s acquisition time.     

The Development of the In Situ Modification of 1st Generation Analytical Scale Silica Monoliths

Analytical scale silica monoliths are commercially limited to three column selectivities (bare silica, C8 and C18). An in situ modification is reported in detail to overcome this barrier and allow for any functionality of choice to be bonded to the silica surface of the monolithic stationary phase support. The modification method was conducted on a commercial bare silica column to bond the C18 moiety to the silica surface through a silylation reaction. The C18 type of stationary phase was chosen, as this is the most commonly bonded functionality for the majority of stationary phases used for high-performance liquid chromatography (HPLC) separations. The C18-modified monolith’s performance was compared to a commercial C18 monolithic and a particle packed column of the same analytical scale column dimensions (100 × 4.6 mm). The modified C18 monolith proved to be of high quality with an efficiency of 73,267 N m^?1, fast analysis times (operated at flow rates up to 3 mL min^?1 using a conventional 400 bar HPLC system) and improved resolution of a set of polar and non-polar substituted aromatics in comparison to a commercial C18 monolith.     

The Number of Accumulated Photons and the Quality of Stimulated Emission Depletion Lifetime Images

Time binning is used to increase the number of photon counts in the peak channel of stimulated emission depletion fluorescence lifetime decay curves to determine how it affects the resulting lifetime image. The fluorescence lifetime of the fluorophore, Alexa Fluor 594 phalloidin, bound to F‐actin is probed in cultured S2 cells at a spatial resolution of ~40 nm. This corresponds to a 10‐fold smaller probe volume compared to confocal imaging, and a reduced number of photons contributing to the signal. Pixel‐by‐pixel fluorescence lifetime measurements and error analysis show that an average of 40 ± 30 photon counts in the peak channel with a signal‐to‐noise ratio of 20 is enough to calculate a reliable fluorescence lifetime from a single exponential fluorescence decay. No heterogeneity in the actin cytoskeleton in different regions of the cultured cells was measured in the 40–400 nm spatial regime.     

A stripe-to-droplet transition driven by conformational transitions in a binary lipid-lipopolymer mixture at the air-water interface

We report the observation of an unusual stripe-droplet transition in precompressed Langmuir monolayers consisting of mixtures of poly(ethylene) glycol (PEG) amphiphiles and phospholipids. This highly reproducible and fully reversible transition occurs at approximately zero surface pressure during expansion (or compression) of the monolayer following initial compression into a two-dimensional solid phase. It is characterized by spontaneous emergence of an extended, disordered stripe-like morphology from an optically homogeneous phase during gradual expansion. These stripe patterns appear as a transient feature and continuously progress, involving gradual coarsening and ultimate transformation into a droplet morphology upon further expansion. Furthermore, varying relative concentrations of the two amphiphiles and utilizing amphiphiles with considerably longer ethylene glycol headgroups reveal that this pattern evolution occurs in narrow concentration regimes, values of which depend on ethylene oxide headgroup size. These morphological transitions are reminiscent of those seen during a passage through a critical point by variations in thermodynamic parameters (e.g., temperature or pressure) as well as those involving spinodal decomposition. While the precise mechanism cannot be ascertained using present experiments alone, our observations can be reconciled in terms of modulations in competing interactions prompted by the pancake-mushroom-brush conformational transitions of the ethylene glycol headgroup. This in turn suggests that the conformational degree of freedom represents an independent order parameter, or a switch, which can induce large-scale structural reorganization in amphiphilic monolayers. Because molecular conformational changes are pervasive in biological membranes, we speculate that such conformational transition-induced pattern evolution might provide a physical mechanism by which membrane processes are amplified.     

Aluminium in the human brain

Abstract  

An inevitable consequence of humans living in the Aluminium Age is the presence of aluminium in the brain. This non-essential, neurotoxic metal gains entry to the brain throughout all stages of human development, from the foetus through to old age. Human exposure to myriad forms of this ubiquitous and omnipresent metal makes its presence in the brain inevitable, while the structure and physiology of the brain makes it particularly susceptible to the accumulation of aluminium with age. In spite of aluminium’s complete lack of biological essentiality, it actually participates avidly in brain biochemistry and substitutes for essential metals in critical biochemical processes. The degree to which such substitutions are disruptive and are manifested as biological effects will depend upon the biological availability of aluminium in any particular physical or chemical compartment, and will under all circumstances be exerting an energy load on the brain. In short, the brain must expend energy in its ‘unconscious’ response to an exposure to biologically available aluminium. There are many examples where ‘biological effect’ has resulted in aluminium-induced neurotoxicity and most potently in conditions that have resulted in an aluminium-associated encephalopathy. However, since aluminium is non-essential and not required by the brain, its biological availability will only rarely achieve such levels of acuity, and it is more pertinent to consider and investigate the brain’s response to much lower though sustained levels of biologically reactive aluminium. This is the level of exposure that defines the putative role of aluminium in chronic neurodegenerative disease and, though thoroughly investigated in numerous animal models, the chronic toxicity of aluminium has yet to be addressed experimentally in humans. A feasible test of the ‘aluminium hypothesis’, whereby aluminium in the human brain is implicated in chronic neurodegenerative disease, would be to reduce the brain’s aluminium load to the lowest possible level by non-invasive means. The simplest way that this aim can be fulfilled in a significant and relevant population is by facilitating the urinary excretion of aluminium through the regular drinking of a silicic acid-rich mineral water over an extended time period. This will lower the body and brain burden of aluminium, and by doing so will test whether brain aluminium contributes significantly to chronic neurodegenerative diseases such as Alzheimer’s and Parkinson’s.