Rui Nian

Extreme learning machine towards dynamic model hypothesis in fish ethology research

In this paper, we present one dynamic model hypothesis to perform fish trajectory tracking in the fish ethology research and develop the relevant mathematical criterion on the basis of the Extreme Learning Machine (ELM). It is shown that the proposed scheme can conduct the non-linear and non Gaussian tracking process by multiple historical cues and current predictions - the state vector motion, the color distribution and the appearance recognition, all of which can be extracted from the single-hidden layer feedforward neural network (SLFN) at diverse levels with ELM. The strategy of the hierarchical hybrid ELM ensemble then combines the individual SLFN of the tracking cues for the performance improvements. The simulation results have shown the excellent performance in both robustness and accuracy of the developed approach.

Fast Face Recognition Via Sparse Coding and Extreme Learning Machine

Most face recognition approaches developed so far regard the sparse coding as one of the essential means, while the sparse coding models have been hampered by the extremely expensive computational cost in the implementation. In this paper, a novel scheme for the fast face recognition is presented via extreme learning machine (ELM) and sparse coding. The common feature hypothesis is first introduced to extract the basis function from the local universal images, and then the single hidden layer feedforward network (SLFN) is established to simulate the sparse coding process for the face images by ELM algorithm. Some developments have been done to maintain the efficient inherent information embedding in the ELM learning. The resulting local sparse coding coefficient will then be grouped into the global representation and further fed into the ELM ensemble which is composed of a number of SLFNs for face recognition. The simulation results have shown the good performance in the proposed approach that could be comparable to the state-of-the-art techniques at a much higher speed.

Nonspecific interactions of chromatin with immunoglobulin G and protein A, and their impact on purification performance

Chromatin released from dead host cells during in vitro production of IgG monoclonal antibodies exists mostly in complex hetero-aggregates consisting of nucleosomal arrays (DNA+histone proteins), non-histone proteins, and aberrant forms of IgG. They bind immobilized protein A more aggressively than IgG, through their nucleosomal histone components, and hinder access of IgG to Fc-specific binding sites, thereby reducing dynamic binding capacity. The majority of host cell contaminants in eluted IgG are leachates from chromatin hetero-aggregates that remain bound to protein A. Formation of turbidity in eluted IgG during pH titration is caused by neutral-pH insolubility of chromatin hetero-aggregates. NaOH is required at 500 mM to remove accumulated chromatin. A chromatin-directed clarification method removed 99% of histones, 90% of non-histone proteins, achieved a 6 log reduction of DNA, 4 log reduction of lipid-enveloped virus, and 5 log reduction of non-enveloped retrovirus, while conserving 98% of the native IgG. This suspended most of performance compromises imposed on protein A. IgG binding capacity increased ~20%. Host protein contamination was reduced about 100-fold compared to protein A loaded with harvest clarified by centrifugation and microfiltration. Aggregates were reduced to less than 0.05%. Turbidity of eluted IgG upon pH neutralization was nearly eliminated. Column cleaning was facilitated by minimizing the accumulation of chromatin.

AUV SLAM and Experiments Using a Mechanical Scanning Forward-Looking Sonar

Navigation technology is one of the most important challenges in the applications of autonomous underwater vehicles (AUVs) which navigate in the complex undersea environment. The ability of localizing a robot and accurately mapping its surroundings simultaneously, namely the simultaneous localization and mapping (SLAM) problem, is a key prerequisite of truly autonomous robots. In this paper, a modified-FastSLAM algorithm is proposed and used in the navigation for our C-Ranger research platform, an open-frame AUV. A mechanical scanning imaging sonar is chosen as the active sensor for the AUV. The modified-FastSLAM implements the update relying on the on-board sensors of C-Ranger. On the other hand, the algorithm employs the data association which combines the single particle maximum likelihood method with modified negative evidence method, and uses the rank-based resampling to overcome the particle depletion problem. In order to verify the feasibility of the proposed methods, both simulation experiments and sea trials for C-Ranger are conducted. The experimental results show the modified-FastSLAM employed for the navigation of the C-Ranger AUV is much more effective and accurate compared with the traditional methods.

Cleavage efficient 2A peptides for high level monoclonal antibody expression in CHO cells

Linking the heavy chain (HC) and light chain (LC) genes required for monoclonal antibodies (mAb) production on a single cassette using 2A peptides allows control of LC and HC ratio and reduces non-expressing cells. Four 2A peptides derived from the foot-and-mouth disease virus (F2A), equine rhinitis A virus (E2A), porcine teschovirus-1 (P2A) and Thosea asigna virus (T2A), respectively, were compared for expression of 3 biosimilar IgG1 mAbs in Chinese hamster ovary (CHO) cell lines. HC and LC were linked by different 2A peptides both in the absence and presence of GSG linkers. Insertion of a furin recognition site upstream of 2A allowed removal of 2A residues that would otherwise be attached to the HC. Different 2A peptides exhibited different cleavage efficiencies that correlated to the mAb expression level. The relative cleavage efficiency of each 2A peptide remains similar for expression of different IgG1 mAbs in different CHO cells. While complete cleavage was not observed for any of the 2A peptides, GSG linkers did enhance the cleavage efficiency and thus the mAb expression level. T2A with the GSG linker (GT2A) exhibited the highest cleavage efficiency and mAb expression level. Stably amplified CHO DG44 pools generated using GT2A had titers 357, 416 and 600 mg/L for the 3 mAbs in shake flask batch cultures. Incomplete cleavage likely resulted in incorrectly processed mAb species and aggregates, which were removed with a chromatin-directed clarification method and protein A purification. The vector and methods presented provide an easy process beneficial for both mAb development and manufacturing.

Ensemble delta test-extreme learning machine (DT-ELM) for regression

Extreme learning machine (ELM) has shown its good performance in regression applications with a very fast speed. But there is still a difficulty to compromise between better generalization performance and smaller complexity of the ELM (a number of hidden nodes). This paper proposes a method called Delta Test-ELM (DT-ELM), which operates in an incremental way to create less complex ELM structures and determines the number of hidden nodes automatically. It uses Bayesian Information Criterion (BIC) as well as Delta Test (DT) to restrict the search as well as to consider the size of the network and prevent overfitting. Moreover, ensemble modeling is used on different DT-ELM models and it shows good test results in Experiments section.

Transient conformational modification of immunoglobulin G during purification by protein A affinity chromatography

Exposure of three native IgG1 monoclonal antibodies to 100mM acetate, pH 3.5 had no significant effect on their hydrodynamic size (11.5±0.5nm), while elution from protein A with the same buffer created a conformation of 5.5±1.0nm. Formation of the reduced-size conformation was preceded by the known destabilization of the second constant domain of the heavy chain (Cγ2) by contact with protein A, then compounded by exposure to low pH, creating extended flexibility in the hinge-Cγ2 region and allowing the Fab region to fold over the Fc region. The reduced-size conformation was necessary for complete elution. It persisted unchanged for at least 7 days under elution conditions. Physiological conditions restored native size, and it was maintained on re-exposure to 100mM acetate, pH 3.5. Protein A-mediated destabilization and subsequent restoration of native size did not create aggregates, but the reduced-size conformation was more susceptible to aggregation by secondary stress than native antibody. Protein A-mediated formation of the reduced-size conformation is probably universal during purification of human IgG1 antibodies, and may occur with other subclasses and IgG from other species, as well as Fc-fusion proteins.

Extreme learning machine for missing data using multiple imputations

In the paper, we examine the general regression problem under the missing data scenario. In order to provide reliable estimates for the regression function (approximation), a novel methodology based on Gaussian Mixture Model and Extreme Learning Machine is developed. Gaussian Mixture Model is used to model the data distribution which is adapted to handle missing values, while Extreme Learning Machine enables to devise a multiple imputation strategy for final estimation. With multiple imputation and ensemble approach over many Extreme Learning Machines, final estimation is improved over the mean imputation performed only once to complete the data. The proposed methodology has longer running times compared to simple methods, but the overall increase in accuracy justifies this trade-off.

A Two-Stage Methodology using K-NN and False Positive Minimizing ELM for Nominal Data Classification

This paper focuses on the problem of making decisions in the context of nominal data under specific constraints. The underlying goal driving the methodology proposed here is to build a decision-making model capable of classifying as many samples as possible while avoiding false positives at all costs, all within the smallest possible computational time. Under such constraints, one of the best type of model is the cognitive-inspired extreme learning machine (ELM), for the final decision process. A two-stage decision methodology using two types of classifiers, a distance-based one, K-NN, and the cognitive-based one, ELM, provides a fast means of obtaining a classification decision on a sample, keeping false positives as low as possible while classifying as many samples as possible (high coverage). The methodology only has two parameters, which, respectively, set the precision of the distance approximation and the final trade-off between false-positive rate and coverage. Experimental results using a specific dataset provided by F-Secure Corporation show that this methodology provides a rapid decision on new samples, with a direct control over the false positives and thus on the decision capabilities of the model.

Chromatographic biopanning for the selection of peptides with high specificity to Pb2+ from phage displayed peptide library.

Toxic heavy metal pollution is a global problem occurring in air, soil as well as water. There is a need for a more cost effective, renewable remediation technique, but most importantly, for a recovery method that is selective for one specific metal of concern. Phage display technology has been used as a powerful tool in the discovery of peptides capable of exhibiting specific affinity to various metals or metal ions. However, traditional phage display is mainly conducted in batch mode, resulting in only one equilibrium state hence low-efficiency selection. It is also unable to monitor the selection process in real time mode. In this study, phage display technique was incorporated with chromatography procedure with the use of a monolithic column, facilitating multiple phage-binding equilibrium states and online monitoring of the selection process in search of affinity peptides to Pb2+. In total, 17 candidate peptides were found and their specificity toward Pb2+ was further investigated with bead-based enzyme immunoassay (EIA). A highly specific Pb2+ binding peptide ThrAsnThrLeuSerAsnAsn (TNTLSNN) was obtained. Based on our knowledge, this is the first report on a new chromatographic biopanning method coupled with monolithic column for the selection of metal ion specific binding peptides. It is expected that this monolith-based chromatographic biopanning will provide a promising approach for a high throughput screening of affinity peptides cognitive of a wide range of target species.