Juan Luis Fernández-Martínez

Stochastic Stability Analysis of the Linear Continuous and Discrete PSO Models

Particle swarm optimization (PSO) can be interpreted physically as a particular discretization of a stochastic damped mass-spring system. Knowledge of this analogy has been crucial to derive the PSO continuous model and to introduce different PSO family members including the generalized PSO (GPSO) algorithm, which is the generalization of PSO for any time discretization step. In this paper, we present the stochastic analysis of the linear continuous and generalized PSO models for the case of a stochastic center of attraction. Analysis of the GPSO second order trajectories is performed and clarifies the roles of the PSO parameters and that of the cost function through the algorithm execution: while the PSO parameters mainly control the eigenvalues of the dynamical systems involved, the mean trajectory of the center of attraction and its covariance functions with the trajectories and their derivatives (or the trajectories in the near past) act as forcing terms to update first and second order trajectories. The similarity between the oscillation center dynamics observed for different kinds of benchmark functions might explain the PSO success for a broad range of optimization problems. Finally, a comparison between real simulations and the linear continuous PSO and GPSO models is shown. As expected, the GPSO tends to the continuous PSO when time step approaches zero. Both models account fairly well for the dynamics (first and second order moments) observed in real runs. This analysis constitutes so far the most realistic attempt to better understand and approach the real PSO dynamics from a stochastic point of view.

Supervised Classification by Filter Methods and Recursive Feature Elimination Predicts Risk of Radiotherapy-Related Fatigue in Patients with Prostate Cancer

Background Fatigue is a common side effect of cancer (CA) treatment. We used a novel analytical method to identify and validate a specific gene cluster that is predictive of fatigue risk in prostate cancer patients (PCP) treated with radiotherapy (RT). Methods A total of 44 PCP were categorized into high-fatigue (HF) and low-fatigue (LF) cohorts based on fatigue score change from baseline to RT completion. Fold-change differential and Fishers linear discriminant analyses (LDA) from 27 subjects with gene expression data at baseline and RT completion generated a reduced base of most discriminatory genes (learning phase). A nearest-neighbor risk (k-NN) prediction model was developed based on small-scale prognostic signatures. The predictive model validity was tested in another 17 subjects using baseline gene expression data (validation phase). Result The model generated in the learning phase predicted HF classification at RT completion in the validation phase with 76.5% accuracy. Conclusion The results suggest that a novel analytical algorithm that incorporates fold-change differential analysis, LDA, and a k-NN may have applicability in predicting regimen-related toxicity in cancer patients with high reliability, if we take into account these results and the limited amount of data that we had at disposal. It is expected that the accuracy will be improved by increasing data sampling in the learning phase.

Adherence to a Mediterranean Diet Influences the Fecal Metabolic Profile of Microbial-Derived Phenolics in a Spanish Cohort of Middle-Age and Older People

Despite the evidence regarding the influence of certain polyphenol food sources on the metabolic profile in feces, the association between the different phenolics provided by the diet and the fecal phenolic profile has not been elucidated. In this study, the composition of phenolic metabolites in fecal solutions was analyzed by UPLC-ESI-MS/MS in 74 volunteers. This fecal phenolic profile showed a high interindividual variation of the different compounds analyzed, phenylacetic and phenylpropionic acids being the major classes of phenolic metabolites excreted in feces. Subjects with higher adherence to a Mediterranean dietary pattern presented greater fecal concentrations of benzoic and 3-hydroxyphenylacetic acids, positively correlated with the intake of the principal classes and subclasses of polyphenols and fibers, and higher levels of Clostridium cluster XVIa and Faecalibacterium prausnitzii. These results provide a link among the Mediterranean dietary pattern, the bioactive compounds of the diet, and the fecal metabolic phenolic profile.

STOCHASTIC STABILITY AND NUMERICAL ANALYSIS OF TWO NOVEL ALGORITHMS OF THE PSO FAMILY: PP-GPSO AND RR-GPSO

The PSO algorithm can be physically interpreted as a stochastic damped mass-spring system: the so-called PSO continuous model. Furthermore, PSO corresponds to a particular discretization of the PSO...The PSO algorithm can be physically interpreted as a stochastic damped mass-spring system: the so-called PSO continuous model. Furthermore, PSO corresponds to a particular discretization of the PSO continuous model. Based on this mechanical analogy we derived in the past a family of PSO-like versions, where the acceleration is discretized using a centered scheme and the velocity of the particles can be regressive (GPSO), progressive (CP-GPSO) or centered (CC-GPSO). Although the first and second order trajectories of these algorithms are isomorphic, CC-GPSO and CP-GPSO are very different from GPSO. In this paper we present two other PSO-like methods: PP-GPSO and RR-GPSO. These algorithms correspond respectively to progressive and regressive discretizations in acceleration and velocity. PP-PSO has the same velocity update than GPSO, but the velocities used to update the trajectories are delayed one iteration, thus, PP-PSO acts as a Jacobi system updating positions and velocities at the same time. RR-GPSO is similar to a GPSO with stochastic constriction factor. Both versions have a very different behavior from GPSO and the other family members introduced in the past: CC-PSO and CP-PSO. RR-PSO seems to have the greatest convergence rate and its good parameter sets can be calculated analytically since they are along a straight line located in the first order stability region. Conversely PP-PSO seems to be a more explorative version, although the behavior of these algorithms can be partly problem dependent. Both exhibit a very peculiar behavior, very different from other family members, and thus they can be called distant PSO relatives. RR-PSO have the greatest convergence rate of all family members for a wide range of benchmark functions with different numerical complexity in 10, 30 and 50 dimensions. These algorithms have been succesfully applied for protein secondary structure prediction and in oil and gas reservoir optimization.

On the prediction of Hodgkin lymphoma treatment response

PurposeThe cure rate in Hodgkin lymphoma is high, but the response along with treatment is still unpredictable and highly variable among patients. Detecting those patients who do not respond to treatment at early stages could bring improvements in their treatment. This research tries to identify the main biological prognostic variables currently gathered at diagnosis and design a simple machine learning methodology to help physicians improve the treatment response assessment.MethodsWe carried out a retrospective analysis of the response to treatment of a cohort of 263 Caucasians who were diagnosed with Hodgkin lymphoma in Asturias (Spain). For that purpose, we used a list of 35 clinical and biological variables that are currently measured at diagnosis before any treatment begins. To establish the list of most discriminatory prognostic variables for treatment response, we designed a machine learning approach based on two different feature selection methods (Fisher’s ratio and maximum percentile distance) and backwards recursive feature elimination using a nearest-neighbor classifier (k-NN). The weights of the k-NN classifier were optimized using different terms of the confusion matrix (true- and false-positive rates) to minimize risk in the decisions.Results and conclusionsWe found that the optimum strategy to predict treatment response in Hodgkin lymphoma consists in solving two different binary classification problems, discriminating first if the patient is in progressive disease; if not, then discerning among complete and partial remission. Serum ferritin turned to be the most discriminatory variable in predicting treatment response, followed by alanine aminotransferase and alkaline phosphatase. The importance of these prognostic variables suggests a close relationship between inflammation, iron overload, liver damage and the extension of the disease.

A Novel Peptide for Simultaneously Enhanced Treatment of Head and Neck Cancer and Mitigation of Oral Mucositis.

We have characterized a novel 21 amino acid-peptide derived from Antrum Mucosal Protein (AMP)-18 that mediates growth promotion of cultured normal epithelial cells and mitigates radiation-induced oral mucositis in animal models, while suppressing in vitro function of cancer cells. The objective of this study was to evaluate these dual potential therapeutic effects of AMP peptide in a clinically relevant animal model of head and neck cancer (HNC) by simultaneously assessing its effect on tumor growth and radiation-induced oral mucositis in an orthotopic model of HNC. Bioluminescent SCC-25 HNC cells were injected into the anterior tongue and tumors that formed were then subjected to focal radiation treatment. Tumor size was assessed using an in vivo imaging system, and the extent of oral mucositis was compared between animals treated with AMP peptide or vehicle (controls). Synergism between AMP peptide and radiation therapy was suggested by the finding that tumors in the AMP peptide/radiation therapy cohort demonstrated inhibited growth vs. radiation therapy-only treated tumors, while AMP peptide-treatment delayed the onset and reduced the severity of radiation therapy-induced oral mucositis. A differential effect on apoptosis appears to be one mechanism by which AMP-18 can stimulate growth and repair of injured mucosal epithelial cells while inhibiting proliferation of HNC cells. RNA microarray analysis identified pathways that are differentially targeted by AMP-18 in HNC vs. nontransformed cells. These observations confirm the notion that normal cells and tumor cells may respond differently to common biological stimuli, and that leveraging this finding in the case of AMP-18 may provide a clinically relevant opportunity.

Exploring the Uncertainty Space of Ensemble Classifiers in Face Recognition

In this paper, we present a supervised ensemble learning algorithm, called SCAV1, and its application to face recognition. This algorithm exploits the uncertainty space of the ensemble classifiers. Its design includes six different nearest-neighbor (NN) classifiers that are based on different and diverse image attributes: histogram, variogram, texture analysis, edges, bidimensional discrete wavelet transform and Zernike moments. In this approach each attribute, together with its corresponding type of the analysis (local or global), and the distance criterion (p-norm) induces a different individual NN classifier. The ensemble classifier SCAV1 depends on a set of parameters: the number of candidate images used by each individual method to perform the final classification and the individual weights given to each individual classifier. SCAV1 parameters are optimized/sampled using a supervised approach via the regressive particle swarm optimization algorithm (RR-PSO). The final classifier exploits the uncertainty space of SCAV1 and uses majority voting (Borda Count) as a final decision rule. We show the application of this algorithm to the ORL and PUT image databases, obtaining very high and stable accuracies (100% median accuracy and almost null interquartile range). In conclusion, exploring the uncertainty space of ensemble classifiers provides optimum results and seems to be the appropriate strategy to adopt for face recognition and other classification problems.

Genomic Data Integration in Chronic Lymphocytic Leukemia

B‐cell chronic lymphocytic leukemia (CLL) is a heterogeneous disease and the most common adult leukemia in western countries. IgVH mutational status distinguishes two major types of CLL, each associated with a different prognosis and survival. Sequencing identified NOTCH1 and SF3B1 as the two main recurrent mutations. We described a novel method to clarify how these mutations affect gene expression by finding small‐scale signatures that predict the IgVH, NOTCH1 and SF3B1 mutations. We subsequently defined the biological pathways and correlation networks involved in disease development, with the potential goal of identifying new drugable targets.

Sensitivity analysis of gene ranking methods in phenotype prediction.

INTRODUCTION It has become clear that noise generated during the assay and analytical processes has the ability to disrupt accurate interpretation of genomic studies. Not only does such noise impact the scientific validity and costs of studies, but when assessed in the context of clinically translatable indications such as phenotype prediction, it can lead to inaccurate conclusions that could ultimately impact patients. We applied a sequence of ranking methods to damp noise associated with microarray outputs, and then tested the utility of the approach in three disease indications using publically available datasets. MATERIALS AND METHODS This study was performed in three phases. We first theoretically analyzed the effect of noise in phenotype prediction problems showing that it can be expressed as a modeling error that partially falsifies the pathways. Secondly, via synthetic modeling, we performed the sensitivity analysis for the main gene ranking methods to different types of noise. Finally, we studied the predictive accuracy of the gene lists provided by these ranking methods in synthetic data and in three different datasets related to cancer, rare and neurodegenerative diseases to better understand the translational aspects of our findings. RESULTS AND DISCUSSION In the case of synthetic modeling, we showed that Fishers Ratio (FR) was the most robust gene ranking method in terms of precision for all the types of noise at different levels. Significance Analysis of Microarrays (SAM) provided slightly lower performance and the rest of the methods (fold change, entropy and maximum percentile distance) were much less precise and accurate. The predictive accuracy of the smallest set of high discriminatory probes was similar for all the methods in the case of Gaussian and Log-Gaussian noise. In the case of class assignment noise, the predictive accuracy of SAM and FR is higher. Finally, for real datasets (Chronic Lymphocytic Leukemia, Inclusion Body Myositis and Amyotrophic Lateral Sclerosis) we found that FR and SAM provided the highest predictive accuracies with the smallest number of genes. Biological pathways were found with an expanded list of genes whose discriminatory power has been established via FR. CONCLUSIONS We have shown that noise in expression data and class assignment partially falsifies the sets of discriminatory probes in phenotype prediction problems. FR and SAM better exploit the principle of parsimony and are able to find subsets with less number of high discriminatory genes. The predictive accuracy and the precision are two different metrics to select the important genes, since in the presence of noise the most predictive genes do not completely coincide with those that are related to the phenotype. Based on the synthetic results, FR and SAM are recommended to unravel the biological pathways that are involved in the disease development.

Relationship of Mitochondrial Enzymes to Fatigue Intensity in Men With Prostate Cancer Receiving External Beam Radiation Therapy

Purpose: Mitochondrial dysfunction is a plausible biological mechanism for cancer-related fatigue. Specific aims of this study were to (1) describe the levels of mitochondrial oxidative phosphorylation complex (MOPC) enzymes, fatigue, and health-related quality of life (HRQOL) before and at completion of external beam radiation therapy (EBRT) in men with nonmetastatic prostate cancer (PC); (2) examine relationships over time among levels of MOPC enzymes, fatigue, and HRQOL; and (3) compare levels of MOPC enzymes in men with clinically significant and nonsignificant fatigue intensification during EBRT. Methods: Fatigue was measured by the revised Piper Fatigue Scale and the Functional Assessment of Cancer Therapy–Fatigue subscale (FACT-F). MOPC enzymes (Complexes I–V) and mitochondrial antioxidant superoxide dismutase 2 were measured in peripheral blood using enzyme-linked immunosorbent assay at baseline and completion of EBRT. Participants were categorized into high or low fatigue (HF vs. LF) intensification groups based on amount of change in FACT-F scores during EBRT. Results: Fatigue reported by the 22 participants with PC significantly worsened and HRQOL significantly declined from baseline to EBRT completion. The HF group comprised 12 men with clinically significant change in fatigue (HF) during EBRT. Although no significant changes were observed in MOPC enzymes from baseline to EBRT completion, there were important differences in the patterns in the levels of MOPC enzymes between HF and LF groups. Conclusion: Distinct patterns of changes in the absorbance of MOPC enzymes delineated fatigue intensification among participants. Further investigation using a larger sample is warranted.