Search

Many machine learning techniques are used as drug discovery tools with the intent to speed characterization by determining relationships between compound structure and biological function. However, particularly in anticancer drug discovery, these models often make only binary decisions about the biological activity for a narrow scope of drug targets. We present a feed-forward neural network, PECAN (Prediction Engine for the Cytostatic Activity of Natural product-like compounds), that simultaneously classifies the potential antiproliferative activity of compounds against 59 cancer cell lines. It predicts the activity to be one of six categories, indicating not only if activity is present but the degree of activity. Using an independent subset of NCI data as a test set, we show that PECAN can reach 60.1% accuracy in a six-way classification and present further evidence that it classifies based on useful structural features of compounds using a “within-one” measure that reaches 93.0% accuracy.

Computational approaches such as genome and metabolome mining are becoming essential to natural products (NPs) research. Consequently, a need exists for an automated structure-type classification system to handle the massive amounts of data appearing for NP structures. An ideal semantic ontology for the classification of NPs should go beyond the simple presence/absence of chemical substructures, but also include the taxonomy of the producing organism, the nature of the biosynthetic pathway, and/or their biological properties. Thus, a holistic and automatic NP classification framework could have considerable value to comprehensively navigate the relatedness of NPs, and especially so when analyzing large numbers of NPs. Here, we introduce NPClassifier, a deep-learning tool for the automated structural classification of NPs from their counted Morgan fingerprints. NPClassifier is expected to accelerate and enhance NP discovery by linking NP structures to their underlying properties.

A thiazole-containing cyclic depsipeptide with 11 amino acid residues, named pagoamide A (1), was isolated from laboratory cultures of a marine Chlorophyte, Derbesiasp. This green algal sample was collected from America Samoa, and pagoamide A was isolated using guidance by MS/MS-based molecular networking. Cultures were grown in a light- and temperature-controlled environment and harvested after several months of growth. The planar structure of pagoamide A (1) was characterized by detailed 1D and 2D NMR experiments along with MS and UV analysis. The absolute configurations of its amino acid residues were determined by advanced Marfey’s analysis following chemical hydrolysis and hydrazinolysis reactions. Two of the residues in pagoamide A (1), phenylalanine and serine, each occurred twice in the molecule, once in the d- and once in the l-configuration. The biosynthetic origin of pagoamide A (1) was considered in light of other natural products investigations with coenocytic green algae.

This report describes the first application of the novel NMR-based machine learning tool “Small Molecule Accurate Recognition Technology” (SMART 2.0) for mixture analysis and subsequent accelerated discovery and characterization of new natural products. The concept was applied to the extract of a filamentous marine cyanobacterium known to be a prolific producer of cytotoxic natural products. This environmental Symplocaextract was roughly fractionated, and then prioritized and guided by cancer cell cytotoxicity, NMR-based SMART 2.0, and MS2-based molecular networking. This led to the isolation and rapid identification of a new chimeric swinholide-like macrolide, symplocolide A, as well as the annotation of swinholide A, samholides A–I, and several new derivatives. The planar structure of symplocolide A was confirmed to be a structural hybrid between swinholide A and luminaolide B by 1D/2D NMR and LC-MS2analysis. A second example applies SMART 2.0 to the characterization of structurally novel cyclic peptides, and compares this approach to the recently appearing “atomic sort” method. This study exemplifies the revolutionary potential of combined traditional and deep learning-assisted analytical approaches to overcome longstanding challenges in natural products drug discovery.

We present a model of the single-ganglion oscillator of the leech swim central pattern generator (CPG). The model is based on the known neuronal architecture of this circuit. Free parameters in the model were fitted to produce membrane potential oscillations matching those seen during swimming. However, the oscillations produced are not robust to small (±5%) changes in the parameters. We propose that this may be due to the large difference between the passive time constant of our model cells and the period of the swim oscillation. We discuss possible ways the real circuit achieves robustness.

We present a thorough analysis of the capabilities of the linear combination (LC) model for fusion of information retrieval systems. The LC model combines the results lists of multiple IR systems by scoring each document using a weighted sum of the scores from each of the component systems. We first present both empirical and analytical justification for the hypotheses that such a model should only be used when the systems involved have high performance, a large overlap of relevant documents, and a small overlap of nonrelevant documents. The empirical approach allows us to very accurately predict the performance of a combined system. We also derive a formula for a theoretically optimal weighting scheme for combining 2 systems. We introduce d—the difference between the average score on relevant documents and the average score on nonrelevant documents—as a performance measure which not only allows mathematical reasoning about system performance, but also allows the selection of weights which generalize well to new documents. We describe a number of experiments involving large numbers of different IR systems which support these findings.

We describe a technique for automatically adapting to the rate of an incoming signal. We first build a model of the signal using a recurrent network trained to predict the input at some delay, for a ‘typical’ rate of the signal. Then, fixing the weights of this network, we adapt the time constant t of the network using gradient descent, adapting the delay appropriately as well. We have found that on simple signals, the network adapts rapidly to new inputs varying in rate from being twice as fast as the original signal, down to ten times as slow. So far our results are based on linear rate changes. We discuss the possibilities of the application of this idea to speech.

Four models were compared on repeated explicit memory (fragment cued recall) or implicit memory (fragment completion) tasks (Hayman & Tulving, 1989a). In the experiments, when given explicit instructions to complete fragments with words from a just-studied list—the explicit condition—people showed a dependence relation between the first and the second fragment targeted at the same word. However, when subjects were just told to complete the (primed) fragments—the implicit condition—stochastic independence between the two fragments resulted. Three distributed models—CHARM, a competitive-learning model, and a back-propagation model produced dependence, as in the explicit memory test. In contrast, a separate-trace model, MINERVA, showed independence, as in the implicit task. It was concluded that explicit memory is based on a highly interactive network that glues or binds together the features within the items, as do the first three models. The binding accounts for the dependence relation. Implicit memory appears to be based, instead, on separate non interacting traces.

• We develop a novel hierarchical reservoir computing (RC) framework called the Deep Projection-encoding Echo State Network (DeePr-ESN) based on projection-encodings between reservoirs, which takes advantage of the merits of reservoir computing and deep learning, and bridges the gap between them. • By unsupervised encoding of echo states layer by layer, the proposed DeePr-ESN can not only provide more robust generalization performance than existing methods, but also obtains more rich multiscale dynamics than other hierarchical RC models. • Compared with the existing Reservoir Computing hierarchical models, the DeePr-ESN achieves better performance on well-known chaotic time series modeling tasks and several real-world time series prediction tasks. Highlights (for review) As a recurrent neural network that requires no training, the reservoir computing (RC) model has attracted widespread attention in the last decade, especially in the context of time series prediction. However, most time series have a multiscale structure, which a single-hidden-layer RC model may have difficulty capturing. In this paper, we propose a novel multiple projection-encoding hierarchical reservoir computing framework called Deep Projection-encoding Echo State Network (DeePr-ESN). The most distinctive feature of our model is its ability to learn multiscale dynamics through stacked ESNs, connected via subspace projections. Specifically, when an input time series is projected into the high-dimensional echo-state space of a reservoir, a subsequent encoding layer (e.g., an autoencoder or PCA) projects the echo-state representations into a lower-dimensional feature space. These representations are the principal components of the echo-state representations, which removes the high frequency components of the representations. These can then be processed by another ESN through random connections. By using projection layers and encoding layers alternately, our DeePr-ESN can provide much more robust generalization performance than previous methods, and also fully takes advantage of the temporal kernel property of ESNs to encode the multiscale dynamics of time series. In our experiments, the DeePr-ESNs outperform both standard ESNs and existing hierarchical reservoir computing models on some artificial and real-world time series prediction tasks. [ABSTRACT FROM AUTHOR]