X-ray microtomography and linear discriminant analysis enable detection of embolism-related acoustic emissions

BACKGROUND

Acoustic emission (AE) sensing is in use since the late 1960s in drought-induced embolism research as a non-invasive and continuous method. It is very well suited to assess a plant's vulnerability to dehydration. Over the last couple of years, AE sensing has further improved due to progress in AE sensors, data acquisition methods and analysis systems. Despite these recent advances, it is still challenging to detect drought-induced embolism events in the AE sources registered by the sensors during dehydration, which sometimes questions the quantitative potential of AE sensing.

RESULTS

In quest of a method to separate embolism-related AE signals from other dehydration-related signals, a 2-year-old potted Fraxinus excelsior L. tree was subjected to a drought experiment. Embolism formation was acoustically measured with two broadband point-contact AE sensors while simultaneously being visualized by X-ray computed microtomography (µCT). A machine learning method was used to link visually detected embolism formation by µCT with corresponding AE signals. Specifically, applying linear discriminant analysis (LDA) on the six AE waveform parameters amplitude, counts, duration, signal strength, absolute energy and partial power in the range 100-200 kHz resulted in an embolism-related acoustic vulnerability curve (VC_AE-E) better resembling the standard µCT VC (VC_CT), both in time and in absolute number of embolized vessels. Interestingly, the unfiltered acoustic vulnerability curve (VC_AE) also closely resembled VC_CT, indicating that VCs constructed from all registered AE signals did not compromise the quantitative interpretation of the species' vulnerability to drought-induced embolism formation.

CONCLUSION

Although machine learning could detect similar numbers of embolism-related AE as µCT, there still is insufficient model-based evidence to conclusively attribute these signals to embolism events. Future research should therefore focus on similar experiments with more in-depth analysis of acoustic waveforms, as well as explore the possibility of Fast Fourier transformation (FFT) to remove non-embolism-related AE signals.

Algebraic shortcuts for leave-one-out cross-validation in supervised network inference

Supervised machine learning techniques have traditionally been very successful at reconstructing biological networks, such as protein-ligand interaction, protein-protein interaction and gene regulatory networks. Many supervised techniques for network prediction use linear models on a possibly nonlinear pairwise feature representation of edges. Recently, much emphasis has been placed on the correct evaluation of such supervised models. It is vital to distinguish between using a model to either predict new interactions in a given network or to predict interactions for a new vertex not present in the original network. This distinction matters because (i) the performance might dramatically differ between the prediction settings and (ii) tuning the model hyperparameters to obtain the best possible model depends on the setting of interest. Specific cross-validation schemes need to be used to assess the performance in such different prediction settings.In this work we discuss a state-of-the-art kernel-based network inference technique called two-step kernel ridge regression. We show that this regression model can be trained efficiently, with a time complexity scaling with the number of vertices rather than the number of edges. Furthermore, this framework leads to a series of cross-validation shortcuts that allow one to rapidly estimate the model performance for any relevant network prediction setting. This allows computational biologists to fully assess the capabilities of their models. The machine learning techniques with the algebraic shortcuts are implemented in the RLScore software package: https://github.com/aatapa/RLScore.

A Comparative Study of Pairwise Learning Methods Based on Kernel Ridge Regression

Many machine learning problems can be formulated as predicting labels for a pair of objects. Problems of that kind are often referred to as pairwise learning, dyadic prediction, or network inference problems. During the past decade, kernel methods have played a dominant role in pairwise learning. They still obtain a state-of-the-art predictive performance, but a theoretical analysis of their behavior has been underexplored in the machine learning literature. In this work we review and unify kernel-based algorithms that are commonly used in different pairwise learning settings, ranging from matrix filtering to zero-shot learning. To this end, we focus on closed-form efficient instantiations of Kronecker kernel ridge regression. We show that independent task kernel ridge regression, two-step kernel ridge regression, and a linear matrix filter arise naturally as a special case of Kronecker kernel ridge regression, implying that all these methods implicitly minimize a squared loss. In addition, we analyze universality, consistency, and spectral filtering properties. Our theoretical results provide valuable insights into assessing the advantages and limitations of existing pairwise learning methods.

miSTAR: miRNA target prediction through modeling quantitative and qualitative miRNA binding site information in a stacked model structure

In microRNA (miRNA) target prediction, typically two levels of information need to be modeled: the number of potential miRNA binding sites present in a target mRNA and the genomic context of each individual site. Single model structures insufficiently cope with this complex training data structure, consisting of feature vectors of unequal length as a consequence of the varying number of miRNA binding sites in different mRNAs. To circumvent this problem, we developed a two-layered, stacked model, in which the influence of binding site context is separately modeled. Using logistic regression and random forests, we applied the stacked model approach to a unique data set of 7990 probed miRNA-mRNA interactions, hereby including the largest number of miRNAs in model training to date. Compared to lower-complexity models, a particular stacked model, named miSTAR (miRNA stacked model target prediction; www.mi-star.org), displays a higher general performance and precision on top scoring predictions. More importantly, our model outperforms published and widely used miRNA target prediction algorithms. Finally, we highlight flaws in cross-validation schemes for evaluation of miRNA target prediction models and adopt a more fair and stringent approach.

Large real-time holographic 3D displays: enabling components and results

A holographic 3D display with 300  mm×200  mm active area was built. The display includes a spatial light modulator that modulates amplitude and phase of light and thus enables holographic reconstruction with high efficiency. Furthermore, holographic optical elements in photopolymer films and laser light sources are used. The requirements on these optical components are discussed. Photographs taken at the display demonstrate that a 3D scene is reconstructed in depth, thus enabling selective accommodation of the observer's eye lenses and natural depth perception. The results demonstrate the advantages of SeeReal's holographic 3D display solution.