A Real-Time Brain-Machine Interface Combining Motor Target and Trajectory Intent Using an Optimal Feedback Control Design

Real-time brain-machine interfaces (BMI) have focused on either estimating the continuous movement trajectory or target intent. However, natural movement often incorporates both. Additionally, BMIs can be modeled as a feedback control system in which the subject modulates the neural activity to move the prosthetic device towards a desired target while receiving real-time sensory feedback of the state of the movement. We develop a novel real-time BMI using an optimal feedback control design that jointly estimates the movement target and trajectory of monkeys in two stages. First, the target is decoded from neural spiking activity before movement initiation. Second, the trajectory is decoded by combining the decoded target with the peri-movement spiking activity using an optimal feedback control design. This design exploits a recursive Bayesian decoder that uses an optimal feedback control model of the sensorimotor system to take into account the intended target location and the sensory feedback in its trajectory estimation from spiking activity. The real-time BMI processes the spiking activity directly using point process modeling. We implement the BMI in experiments consisting of an instructed-delay center-out task in which monkeys are presented with a target location on the screen during a delay period and then have to move a cursor to it without touching the incorrect targets. We show that the two-stage BMI performs more accurately than either stage alone. Correct target prediction can compensate for inaccurate trajectory estimation and vice versa. The optimal feedback control design also results in trajectories that are smoother and have lower estimation error. The two-stage decoder also performs better than linear regression approaches in offline cross-validation analyses. Our results demonstrate the advantage of a BMI design that jointly estimates the target and trajectory of movement and more closely mimics the sensorimotor control system.     

Combining Physicochemical and Evolutionary Information for Protein Contact Prediction

We introduce a novel contact prediction method that achieves high prediction accuracy by combining evolutionary and physicochemical information about native contacts. We obtain evolutionary information from multiple-sequence alignments and physicochemical information from predicted ab initio protein structures. These structures represent low-energy states in an energy landscape and thus capture the physicochemical information encoded in the energy function. Such low-energy structures are likely to contain native contacts, even if their overall fold is not native. To differentiate native from non-native contacts in those structures, we develop a graph-based representation of the structural context of contacts. We then use this representation to train an support vector machine classifier to identify most likely native contacts in otherwise non-native structures. The resulting contact predictions are highly accurate. As a result of combining two sources of information—evolutionary and physicochemical—we maintain prediction accuracy even when only few sequence homologs are present. We show that the predicted contacts help to improve ab initio structure prediction. A web service is available at http://compbio.robotics.tu-berlin.de/epc-map/.     

Eye-Hand Synergy and Intermittent Behaviors during Target-Directed Tracking with Visual and Non-visual Information

Visual feedback and non-visual information play different roles in tracking of an external target. This study explored the respective roles of the visual and non-visual information in eleven healthy volunteers who coupled the manual cursor to a rhythmically moving target of 0.5 Hz under three sensorimotor conditions: eye-alone tracking (EA), eye-hand tracking with visual feedback of manual outputs (EH tracking), and the same tracking without such feedback (EHM tracking). Tracking error, kinematic variables, and movement intermittency (saccade and speed pulse) were contrasted among tracking conditions. The results showed that EHM tracking exhibited larger pursuit gain, less tracking error, and less movement intermittency for the ocular plant than EA tracking. With the vision of manual cursor, EH tracking achieved superior tracking congruency of the ocular and manual effectors with smaller movement intermittency than EHM tracking, except that the rate precision of manual action was similar for both types of tracking. The present study demonstrated that visibility of manual consequences altered mutual relationships between movement intermittency and tracking error. The speed pulse metrics of manual output were linked to ocular tracking error, and saccade events were time-locked to the positional error of manual tracking during EH tracking. In conclusion, peripheral non-visual information is critical to smooth pursuit characteristics and rate control of rhythmic manual tracking. Visual information adds to eye-hand synchrony, underlying improved amplitude control and elaborate error interpretation during oculo-manual tracking.     

Changes in the amplitude and direction of goal-directed hand movements in the lack of visual information

The goal of the present investigation was to determine the precision of goal-directed hand movements in the lack of visual information. The movement amplitude and direction was examined under different experimental conditions. Subjects were ten female and ten male university students. The motor test was drawing 10 cm long straight line and 24 cm long zigzag line in four different experimental conditions. 1) The drawing with open eyes was followed immediately with drawing with closed eyes. 2) The drawing was executed from memory in the lack of visual information. 3) Drawing with restricted amplitude or direction. 4) Drawing with verbal feedback. The errors of the target distance and the lateral deviations from the target were different under the different experimental conditions. The largest errors and underestimation of the target distance occurred in drawing horizontal straight line with closed eyes. No statistically significant gender differences were found. It is concluded that the practice, adjustment of single movement parameter to the target, and the verbal feedback assist better the accuracy of unseen goal-directed hand movement than the recent visual memory.     

HYPROSP: a hybrid protein secondary structure prediction algorithm--a knowledge-based approach

We develop a knowledge-based approach (called PROSP) for protein secondary structure prediction. The knowledge base contains small peptide fragments together with their secondary structural information. A quantitative measure M, called match rate, is defined to measure the amount of structural information that a target protein can extract from the knowledge base. Our experimental results show that proteins with a higher match rate will likely be predicted more accurately based on PROSP. That is, there is roughly a monotone correlation between the prediction accuracy and the amount of structure matching with the knowledge base. To fully utilize the strength of our knowledge base, a hybrid prediction method is proposed as follows: if the match rate of a target protein is at least 80%, we use the extracted information to make the prediction; otherwise, we adopt a popular machine-learning approach. This comprises our hybrid protein structure prediction (HYPROSP) approach. We use the DSSP and EVA data as our datasets and PSIPRED as our underlying machine-learning algorithm. For target proteins with match rate at least 80%, the average Q₃ of PROSP is 3.96 and 7.2 better than that of PSIPRED on DSSP and EVA data, respectively.     

Prediction of MHC Class I Binding Peptides by a Query Learning Algorithm Based on Hidden Markov Models

A query learning algorithm based on hidden Markov models (HMMs) isdeveloped to design experiments for string analysis and prediction of MHCclass I binding peptides. Query learning is introduced to aim at reducingthe number of peptide binding data for training of HMMs. A multiple numberof HMMs, which will collectively serve as a committee, are trained withbinding data and used for prediction in real-number values. The universeof peptides is randomly sampled and subjected to judgement by the HMMs.Peptides whose prediction is least consistent among committee HMMs aretested by experiment. By iterating the feedback cycle of computationalanalysis and experiment the most wanted information is effectivelyextracted. After 7 rounds of active learning with 181 peptides in all,predictive performance of the algorithm surpassed the so far bestperforming matrix based prediction. Moreover, by combining the bothmethods binder peptides (log Kd < -6) could be predicted with84% accuracy. Parameter distribution of the HMMs that can be inspectedvisually after training further offers a glimpse of dynamic specificity ofthe MHC molecules.     

Operon prediction in Pyrococcus furiosus

Identification of operons in the hyperthermophilic archaeon Pyrococcus furiosus represents an important step to understanding the regulatory mechanisms that enable the organism to adapt and thrive in extreme environments. We have predicted operons in P.furiosus by combining the results from three existing algorithms using a neural network (NN). These algorithms use intergenic distances, phylogenetic profiles, functional categories and gene-order conservation in their operon prediction. Our method takes as inputs the confidence scores of the three programs, and outputs a prediction of whether adjacent genes on the same strand belong to the same operon. In addition, we have applied Gene Ontology (GO) and KEGG pathway information to improve the accuracy of our algorithm. The parameters of this NN predictor are trained on a subset of all experimentally verified operon gene pairs of Bacillus subtilis. It subsequently achieved 86.5% prediction accuracy when applied to a subset of gene pairs for Escherichia coli, which is substantially better than any of the three prediction programs. Using this new algorithm, we predicted 470 operons in the P.furiosus genome. Of these, 349 were validated using DNA microarray data.     

Review: protein secondary structure prediction continues to rise

Methods predicting protein secondary structure improved substantially in the 1990s through the use of evolutionary information taken from the divergence of proteins in the same structural family. Recently, the evolutionary information resulting from improved searches and larger databases has again boosted prediction accuracy by more than four percentage points to its current height of around 76% of all residues predicted correctly in one of the three states, helix, strand, and other. The past year also brought successful new concepts to the field. These new methods may be particularly interesting in light of the improvements achieved through simple combining of existing methods. Divergent evolutionary profiles contain enough information not only to substantially improve prediction accuracy, but also to correctly predict long stretches of identical residues observed in alternative secondary structure states depending on nonlocal conditions. An example is a method automatically identifying structural switches and thus finding a remarkable connection between predicted secondary structure and aspects of function. Secondary structure predictions are increasingly becoming the work horse for numerous methods aimed at predicting protein structure and function. Is the recent increase in accuracy significant enough to make predictions even more useful? Because the recent improvement yields a better prediction of segments, and in particular of beta strands, I believe the answer is affirmative. What is the limit of prediction accuracy? We shall see.     

Redundant information encoding in primary motor cortex during natural and prosthetic motor control

Redundant encoding of information facilitates reliable distributed information processing. To explore this hypothesis in the motor system, we applied concepts from information theory to quantify the redundancy of movement-related information encoded in the macaque primary motor cortex (M1) during natural and neuroprosthetic control. Two macaque monkeys were trained to perform a delay center-out reaching task controlling a computer cursor under natural arm movement (manual control, ‘MC’), and using a brain-machine interface (BMI) via volitional control of neural ensemble activity (brain control, ‘BC’). During MC, we found neurons in contralateral M1 to contain higher and more redundant information about target direction than ipsilateral M1 neurons, consistent with the laterality of movement control. During BC, we found that the M1 neurons directly incorporated into the BMI (‘direct’ neurons) contained the highest and most redundant target information compared to neurons that were not incorporated into the BMI (‘indirect’ neurons). This effect was even more significant when comparing to M1 neurons of the opposite hemisphere. Interestingly, when we retrained the BMI to use ipsilateral M1 activity, we found that these neurons were more redundant and contained higher information than contralateral M1 neurons, even though ensembles from this hemisphere were previously less redundant during natural arm movement. These results indicate that ensembles most associated to movement contain highest redundancy and information encoding, which suggests a role for redundancy in proficient natural and prosthetic motor control.     

The role of observers' gaze behaviour when watching object manipulation tasks: predicting and evaluating the consequences of action

When watching an actor manipulate objects, observers, like the actor, naturally direct their gaze to each object as the hand approaches and typically maintain gaze on the object until the hand departs. Here, we probed the function of observers'' eye movements, focusing on two possibilities: (i) that observers'' gaze behaviour arises from processes involved in the prediction of the target object of the actor''s reaching movement and (ii) that this gaze behaviour supports the evaluation of mechanical events that arise from interactions between the actor''s hand and objects. Observers watched an actor reach for and lift one of two presented objects. The observers'' task was either to predict the target object or judge its weight. Proactive gaze behaviour, similar to that seen in self-guided action–observation, was seen in the weight judgement task, which requires evaluating mechanical events associated with lifting, but not in the target prediction task. We submit that an important function of gaze behaviour in self-guided action observation is the evaluation of mechanical events associated with interactions between the hand and object. By comparing predicted and actual mechanical events, observers, like actors, can gain knowledge about the world, including information about objects they may subsequently act upon.     

Prediction of Beta-Turn in Protein Using E-SSpred and Support Vector Machine

β-Turn is a secondary protein structure type that plays an important role in protein configuration and function. Here, we introduced an approach of β-turn prediction that used the support vector machine (SVM) algorithm combined with predicted secondary structure information. The secondary structure information was obtained by using E-SSpred, a new secondary protein structure prediction method. A 7-fold cross validation based on the benchmark dataset of 426 non-homologous protein chains was used to evaluate the performance of our method. The prediction results broke the 80% Q _total barrier and achieved Q _total = 80.9%, MCC = 0.44, and Q _predicted higher 0.9% when compared with the best method. The results in our research are coincident with the conclusion that β-turn prediction accuracy can be improved by inclusion of secondary structure information.     

A Consensus Data Mining secondary structure prediction by combining GOR V and Fragment Database Mining

The major aim of tertiary structure prediction is to obtain protein models with the highest possible accuracy. Fold recognition, homology modeling, and de novo prediction methods typically use predicted secondary structures as input, and all of these methods may significantly benefit from more accurate secondary structure predictions. Although there are many different secondary structure prediction methods available in the literature, their cross-validated prediction accuracy is generally <80%. In order to increase the prediction accuracy, we developed a novel hybrid algorithm called Consensus Data Mining (CDM) that combines our two previous successful methods: (1) Fragment Database Mining (FDM), which exploits the Protein Data Bank structures, and (2) GOR V, which is based on information theory, Bayesian statistics, and multiple sequence alignments (MSA). In CDM, the target sequence is dissected into smaller fragments that are compared with fragments obtained from related sequences in the PDB. For fragments with a sequence identity above a certain sequence identity threshold, the FDM method is applied for the prediction. The remainder of the fragments are predicted by GOR V. The results of the CDM are provided as a function of the upper sequence identities of aligned fragments and the sequence identity threshold. We observe that the value 50% is the optimum sequence identity threshold, and that the accuracy of the CDM method measured by Q(3) ranges from 67.5% to 93.2%, depending on the availability of known structural fragments with sufficiently high sequence identity. As the Protein Data Bank grows, it is anticipated that this consensus method will improve because it will rely more upon the structural fragments.     

Improved flower pollination algorithm for identifying essential proteins

Background

Essential proteins are necessary for the survival and development of cells. The identification of essential proteins can help to understand the minimal requirements for cellular life and it also plays an important role in the disease genes study and drug design. With the development of high-throughput techniques, a large amount of protein-protein interactions data is available to predict essential proteins at the network level. Hitherto, even though a number of essential protein discovery methods have been proposed, the prediction precision still needs to be improved.

Methods

In this paper, we propose a new algorithm, improved Flower Pollination algorithm (FPA) for identifying Essential proteins, named FPE. Different from other existing essential protein discovery methods, we apply FPA which is a new intelligent algorithm imitating pollination behavior of flowering plants in nature to identify essential proteins. Analogous to flower pollination is to find optimal reproduction from the perspective of biological evolution, and the identification of essential proteins is to discover a candidate essential protein set by analyzing the corresponding relationships between FPA algorithm and the prediction of essential proteins, and redefining the positions of flowers and specific pollination process. Moreover, it has been proved that the integration of biological and topological properties can get improved precision for identifying essential proteins. Consequently, we develop a GSC measurement in order to judge the essentiality of proteins, which takes into account not only the Gene expression data, Subcellular localization and protein Complexes information, but also the network topology.

Results

The experimental results show that FPE performs better than the state-of-the-art methods (DC, SC, IC, EC, LAC, NC, PeC, WDC, UDoNC and SON) in terms of the prediction precision, precision-recall curve and jackknife curve for identifying essential proteins and also has high stability.

Conclusions

We confirm that FPE can be used to effectively identify essential proteins by the use of nature-inspired algorithm FPA and the combination of network topology with gene expression data, subcellular localization and protein complexes information. The experimental results have shown the superiority of FPE for the prediction of essential proteins.

     

NRLMFβ: Beta-distribution-rescored neighborhood regularized logistic matrix factorization for improving the performance of drug–target interaction prediction

Techniques for predicting interactions between a drug and a target (protein) are useful for strategic drug repositioning. Neighborhood regularized logistic matrix factorization (NRLMF) is one of the state-of-the-art drug–target interaction prediction methods; it is based on a statistical model using the Bernoulli distribution. However, the prediction is not accurate when drug–target interaction pairs have less interaction information (e.g., the sum of the number of ligands for a target and the number of target proteins for a drug). This study aimed to address this issue by proposing NRLMF with beta distribution rescoring (NRLMFβ), which is an algorithm to improve the score of NRLMF. The score of NRLMFβ is equivalent to the value of the original NRLMF score when the concentration of the beta distribution becomes infinity. The beta distribution is known as a conjugative prior distribution of the Bernoulli distribution and can reflect the amount of interaction information to its shape based on Bayesian inference. Therefore, in NRLMFβ, the beta distribution was used for rescoring the NRLMF score. In the evaluation experiment, we measured the average values of area under the receiver operating characteristics and area under precision versus recall and the 95% confidence intervals. The performance of NRLMFβ was found to be better than that of NRLMF in the four types of benchmark datasets. Thus, we concluded that NRLMFβ improved the prediction accuracy of NRLMF. The source code is available at https://github.com/akiyamalab/NRLMFb.     

Combining evolutionary and structural information for local protein structure prediction

We study the effects of various factors in representing and combining evolutionary and structural information for local protein structural prediction based on fragment selection. We prepare databases of fragments from a set of non-redundant protein domains. For each fragment, evolutionary information is derived from homologous sequences and represented as estimated effective counts and frequencies of amino acids (evolutionary frequencies) at each position. Position-specific amino acid preferences called structural frequencies are derived from statistical analysis of discrete local structural environments in database structures. Our method for local structure prediction is based on ranking and selecting database fragments that are most similar to a target fragment. Using secondary structure type as a local structural property, we test our method in a number of settings. The major findings are: (1) the COMPASS-type scoring function for fragment similarity comparison gives better prediction accuracy than three other tested scoring functions for profile-profile comparison. We show that the COMPASS-type scoring function can be derived both in the probabilistic framework and in the framework of statistical potentials. (2) Using the evolutionary frequencies of database fragments gives better prediction accuracy than using structural frequencies. (3) Finer definition of local environments, such as including more side-chain solvent accessibility classes and considering the backbone conformations of neighboring residues, gives increasingly better prediction accuracy using structural frequencies. (4) Combining evolutionary and structural frequencies of database fragments, either in a linear fashion or using a pseudocount mixture formula, results in improvement of prediction accuracy. Combination at the log-odds score level is not as effective as combination at the frequency level. This suggests that there might be better ways of combining sequence and structural information than the commonly used linear combination of log-odds scores. Our method of fragment selection and frequency combination gives reasonable results of secondary structure prediction tested on 56 CASP5 targets (average SOV score 0.77), suggesting that it is a valid method for local protein structure prediction. Mixture of predicted structural frequencies and evolutionary frequencies improve the quality of local profile-to-profile alignment by COMPASS.     

Protein Function Prediction: From Traditional Classifier to Deep Learning

Deep learning demonstrates greater competence over traditional machine learning techniques for many tasks. In last several years, deep learning has been applied to protein function prediction and a series of good achievements has been obtained. These findings extensively advanced our understanding of protein function. However, the accuracy of protein function prediction based upon deep learning still has yet to be improved. In article number 1900019, Issue 12, Zhang et al. construct DeepFunc, a deep learning framework using derived feature information of protein sequence and protein interactions network. They find that implementing DeepFunc for protein function prediction is more accurate than using DeepGO, a similar method reported previously. Meanwhile, they find that the method of combining multiple derived feature information in DeepFunc is much better than the method of using only single derived feature information. Due to its fully exploiting feature representation learning ability, deep learning with more derived feature information will enable it to be a promising method for solving more complicated protein function prediction problems and other bioinformatics challenges. Recent researches have provided some major insights into the value for using deep learning to protein function prediction problem.     

Unscented Kalman Filter for Brain-Machine Interfaces

Brain machine interfaces (BMIs) are devices that convert neural signals into commands to directly control artificial actuators, such as limb prostheses. Previous real-time methods applied to decoding behavioral commands from the activity of populations of neurons have generally relied upon linear models of neural tuning and were limited in the way they used the abundant statistical information contained in the movement profiles of motor tasks. Here, we propose an n-th order unscented Kalman filter which implements two key features: (1) use of a non-linear (quadratic) model of neural tuning which describes neural activity significantly better than commonly-used linear tuning models, and (2) augmentation of the movement state variables with a history of n-1 recent states, which improves prediction of the desired command even before incorporating neural activity information and allows the tuning model to capture relationships between neural activity and movement at multiple time offsets simultaneously. This new filter was tested in BMI experiments in which rhesus monkeys used their cortical activity, recorded through chronically implanted multielectrode arrays, to directly control computer cursors. The 10th order unscented Kalman filter outperformed the standard Kalman filter and the Wiener filter in both off-line reconstruction of movement trajectories and real-time, closed-loop BMI operation.     

Advantages and limitations of multiple-trait genomic prediction for <Emphasis Type="Italic">Fusarium</Emphasis> head blight severity in hybrid wheat (<Emphasis Type="Italic">Triticum aestivum</Emphasis> L.)

Key message

Predictabilities for wheat hybrids less related to the estimation set were improved by shifting from single- to multiple-trait genomic prediction of Fusarium head blight severity.

Abstract

Breeding for improved Fusarium head blight resistance (FHBr) of wheat is a very laborious and expensive task. FHBr complexity is mainly due to its highly polygenic nature and because FHB severity (FHBs) is greatly influenced by the environment. Associated traits plant height and heading date may provide additional information related to FHBr, but this is ignored in single-trait genomic prediction (STGP). The aim of our study was to explore the benefits in predictabilities of multiple-trait genomic prediction (MTGP) over STGP of target trait FHBs in a population of 1604 wheat hybrids using information on 17,372 single nucleotide polymorphism markers along with indicator traits plant height and heading date. The additive inheritance of FHBs allowed accurate hybrid performance predictions using information on general combining abilities or average performance of both parents without the need of markers. Information on molecular markers and indicator trait(s) improved FHBs predictabilities for hybrids less related to the estimation set. Indicator traits must be observed on the predicted individuals to benefit from MTGP. Magnitudes of genetic and phenotypic correlations along with improvements in predictabilities made plant height a better indicator trait for FHBs than heading date. Thus, MTGP having only plant height as indicator trait already maximized FHBs predictabilities. Provided a good indicator trait was available, MTGP could reduce the impacts of genotype environment \(\times\) interaction on STGP for hybrids less related to the estimation set.

     

The Impact of Augmented Information on Visuo-Motor Adaptation in Younger and Older Adults

Background

Adjustment to a visuo-motor rotation is known to be affected by ageing. According to previous studies, the age-related differences primarily pertain to the use of strategic corrections and the generation of explicit knowledge on which strategic corrections are based, whereas the acquisition of an (implicit) internal model of the novel visuo-motor transformation is unaffected. The present study aimed to assess the impact of augmented information on the age-related variation of visuo-motor adjustments.

Methodology/Principal Findings

Participants performed aiming movements controlling a cursor on a computer screen. Visual feedback of direction of cursor motion was rotated 75° relative to the direction of hand motion. Participants had to adjust to this rotation in the presence and absence of an additional hand-movement target that explicitly depicted the input-output relations of the visuo-motor transformation. An extensive set of tests was employed in order to disentangle the contributions of different processes to visuo-motor adjustment. Results show that the augmented information failed to affect the age-related variations of explicit knowledge, adaptive shifts, and aftereffects in a substantial way, whereas it clearly affected initial direction errors during practice and proprioceptive realignment.

Conclusions

Contrary to expectations, older participants apparently made no use of the augmented information, whereas younger participants used the additional movement target to reduce initial direction errors early during practice. However, after a first block of trials errors increased, indicating a neglect of the augmented information, and only slowly declined thereafter. A hypothetical dual-task account of these findings is discussed. The use of the augmented information also led to a selective impairment of proprioceptive realignment in the younger group. The mere finding of proprioceptive realignment in adaptation to a visuo-motor rotation in a computer-controlled setup is noteworthy since visual and proprioceptive information pertain to different objects.