Implementing statistical learning methods through Bayesian networks (Part 2): bayesian evaluations for results of black toner analyses in forensic document examination

This paper presents and discusses the use of Bayesian procedures - introduced through the use of Bayesian networks in Part I of this series of papers - for 'learning' probabilities from data. The discussion will relate to a set of real data on characteristics of black toners commonly used in printing and copying devices. Particular attention is drawn to the incorporation of the proposed procedures as an integral part in probabilistic inference schemes (notably in the form of Bayesian networks) that are intended to address uncertainties related to particular propositions of interest (e.g., whether or not a sample originates from a particular source). The conceptual tenets of the proposed methodologies are presented along with aspects of their practical implementation using currently available Bayesian network software.

Learning spatial filters for multispectral image segmentation

We present a novel filtering method for multispectral satellite image classification. The proposed method learns a set of spatial filters that maximize class separability of binary support vector machine (SVM) through a gradient descent approach. Regularization issues are discussed in detail and a Frobenius-norm regularization is proposed to efficiently exclude uninformative filters coefficients. Experiments carried out on multiclass one-against-all classification and target detection show the capabilities of the learned spatial filters.

Stochastic stability and the evolution of coordination in spatially structured populations.

Animals can often coordinate their actions to achieve mutually beneficial outcomes. However, this can result in a social dilemma when uncertainty about the behavior of partners creates multiple fitness peaks. Strategies that minimize risk ("risk dominant") instead of maximizing reward ("payoff dominant") are favored in economic models when individuals learn behaviors that increase their payoffs. Specifically, such strategies are shown to be "stochastically stable" (a refinement of evolutionary stability). Here, we extend the notion of stochastic stability to biological models of continuous phenotypes at a mutation-selection-drift balance. This allows us to make a unique prediction for long-term evolution in games with multiple equilibria. We show how genetic relatedness due to limited dispersal and scaled to account for local competition can crucially affect the stochastically-stable outcome of coordination games. We find that positive relatedness (weak local competition) increases the chance the payoff dominant strategy is stochastically stable, even when it is not risk dominant. Conversely, negative relatedness (strong local competition) increases the chance that strategies evolve that are neither payoff nor risk dominant. Extending our results to large multiplayer coordination games we find that negative relatedness can create competition so extreme that the game effectively changes to a hawk-dove game and a stochastically stable polymorphism between the alternative strategies evolves. These results demonstrate the usefulness of stochastic stability in characterizing long-term evolution of continuous phenotypes: the outcomes of multiplayer games can be reduced to the generic equilibria of two-player games and the effect of spatial structure can be analyzed readily.

Decoding sequence learning from single-trial intracranial EEG in humans.

We propose and validate a multivariate classification algorithm for characterizing changes in human intracranial electroencephalographic data (iEEG) after learning motor sequences. The algorithm is based on a Hidden Markov Model (HMM) that captures spatio-temporal properties of the iEEG at the level of single trials. Continuous intracranial iEEG was acquired during two sessions (one before and one after a night of sleep) in two patients with depth electrodes implanted in several brain areas. They performed a visuomotor sequence (serial reaction time task, SRTT) using the fingers of their non-dominant hand. Our results show that the decoding algorithm correctly classified single iEEG trials from the trained sequence as belonging to either the initial training phase (day 1, before sleep) or a later consolidated phase (day 2, after sleep), whereas it failed to do so for trials belonging to a control condition (pseudo-random sequence). Accurate single-trial classification was achieved by taking advantage of the distributed pattern of neural activity. However, across all the contacts the hippocampus contributed most significantly to the classification accuracy for both patients, and one fronto-striatal contact for one patient. Together, these human intracranial findings demonstrate that a multivariate decoding approach can detect learning-related changes at the level of single-trial iEEG. Because it allows an unbiased identification of brain sites contributing to a behavioral effect (or experimental condition) at the level of single subject, this approach could be usefully applied to assess the neural correlates of other complex cognitive functions in patients implanted with multiple electrodes.

A repeated GGA motif is critical for the activity and stability of the riboregulator RsmY of Pseudomonas fluorescens.

The riboregulator RsmY of Pseudomonas fluorescens strain CHA0 is an example of small regulatory RNAs belonging to the global Rsm/Csr regulatory systems controlling diverse cellular processes such as glycogen accumulation, motility, or formation of extracellular products in various bacteria. By binding multiple molecules of the small regulatory protein RsmA, RsmY relieves the negative effect of RsmA on the translation of several target genes involved in the biocontrol properties of strain CHA0. RsmY and functionally related riboregulators have repeated GGA motifs predicted to be exposed in single-stranded regions, notably in the loops of hairpins. The secondary structure of RsmY was corroborated by in vivo cleavage with lead acetate. RsmY mutants lacking three or five (out of six) of the GGA motifs showed reduced ability to derepress the expression of target genes in vivo and failed to bind the RsmA protein efficiently in vitro. The absence of GGA motifs in RsmY mutants resulted in reduced abundance of these transcripts and in a shorter half-life (< or = 6 min as compared with 27 min for wild type RsmY). These results suggest that both the interaction of RsmY with RsmA and the stability of RsmY strongly depend on the GGA repeats and that the ability of RsmY to interact with small regulatory proteins such as RsmA may protect this RNA from degradation.

The control of microtubule stability in vitro and in transfected cells by MAP1B and SCG10.

In neurons, the regulation of microtubules plays an important role for neurite outgrowth, axonal elongation, and growth cone steering. SCG10 family proteins are the only known neuronal proteins that have a strong destabilizing effect, are highly enriched in growth cones and are thought to play an important role during axonal elongation. MAP1B, a microtubule-stabilizing protein, is found in growth cones as well, therefore it was important to test their effect on microtubules in the presence of both proteins. We used recombinant proteins in microtubule assembly assays and in transfected COS-7 cells to analyze their combined effects in vitro and in living cells, respectively. Individually, both proteins showed their expected activities in microtubule stabilization and destruction respectively. In MAP1B/SCG10 double-transfected cells, MAP1B could not protect microtubules from SCG10-induced disassembly in most cells, in particular not in cells that contained high levels of SCG10. This suggests that SCG10 is more potent to destabilize microtubules than MAP1B to rescue them. In microtubule assembly assays, MAP1B promoted microtubule formation at a ratio of 1 MAP1B per 70 tubulin dimers while a ratio of 1 SCG10 per two tubulin dimers was needed to destroy microtubules. In addition to its known binding to tubulin dimers, SCG10 binds also to purified microtubules in growth cones of dorsal root ganglion neurons in culture. In conclusion, neuronal microtubules are regulated by antagonistic effects of MAP1B and SCG10 and a fine tuning of the balance of these proteins may be critical for the regulation of microtubule dynamics in growth cones.

Predicting clinical scores from magnetic resonance scans in Alzheimer's disease.

Machine learning and pattern recognition methods have been used to diagnose Alzheimer's disease (AD) and mild cognitive impairment (MCI) from individual MRI scans. Another application of such methods is to predict clinical scores from individual scans. Using relevance vector regression (RVR), we predicted individuals' performances on established tests from their MRI T1 weighted image in two independent data sets. From Mayo Clinic, 73 probable AD patients and 91 cognitively normal (CN) controls completed the Mini-Mental State Examination (MMSE), Dementia Rating Scale (DRS), and Auditory Verbal Learning Test (AVLT) within 3months of their scan. Baseline MRI's from the Alzheimer's disease Neuroimaging Initiative (ADNI) comprised the other data set; 113 AD, 351 MCI, and 122 CN subjects completed the MMSE and Alzheimer's Disease Assessment Scale-Cognitive subtest (ADAS-cog) and 39 AD, 92 MCI, and 32 CN ADNI subjects completed MMSE, ADAS-cog, and AVLT. Predicted and actual clinical scores were highly correlated for the MMSE, DRS, and ADAS-cog tests (P<0.0001). Training with one data set and testing with another demonstrated stability between data sets. DRS, MMSE, and ADAS-Cog correlated better than AVLT with whole brain grey matter changes associated with AD. This result underscores their utility for screening and tracking disease. RVR offers a novel way to measure interactions between structural changes and neuropsychological tests beyond that of univariate methods. In clinical practice, we envision using RVR to aid in diagnosis and predict clinical outcome.

The influence of learning on evolution: a mathematical framework.

The Baldwin effect can be observed if phenotypic learning influences the evolutionary fitness of individuals, which can in turn accelerate or decelerate evolutionary change. Evidence for both learning-induced acceleration and deceleration can be found in the literature. Although the results for both outcomes were supported by specific mathematical or simulation models, no general predictions have been achieved so far. Here we propose a general framework to predict whether evolution benefits from learning or not. It is formulated in terms of the gain function, which quantifies the proportional change of fitness due to learning depending on the genotype value. With an inductive proof we show that a positive gain-function derivative implies that learning accelerates evolution, and a negative one implies deceleration under the condition that the population is distributed on a monotonic part of the fitness landscape. We show that the gain-function framework explains the results of several specific simulation models. We also use the gain-function framework to shed some light on the results of a recent biological experiment with fruit flies.

Do orthopaedic shoes improve local dynamic stability of gait? An observational study in patients with chronic foot and ankle injuries.

BACKGROUND: Complex foot and ankle fractures, such as calcaneum fractures or Lisfranc dislocations, are often associated with a poor outcome, especially in terms of gait capacity. Indeed, degenerative changes often lead to chronic pain and chronic functional limitations. Prescription footwear represents an important therapeutic tool during the rehabilitation process. Local Dynamic Stability (LDS) is the ability of locomotor system to maintain continuous walking by accommodating small perturbations that occur naturally during walking. Because it reflects the degree of control over the gait, LDS has been advocated as a relevant indicator for evaluating different conditions and pathologies. The aim of this study was to analyze changes in LDS induced by orthopaedic shoes in patients with persistent foot and ankle injuries. We hypothesised that footwear adaptation might help patients to improve gait control, which could lead to higher LDS: METHODS: Twenty-five middle-aged inpatients (5 females, 20 males) participated in the study. They were treated for chronic post-traumatic disabilities following ankle and/or foot fractures in a Swiss rehabilitation clinic. During their stay, included inpatients received orthopaedic shoes with custom-made orthoses (insoles). They performed two 30s walking trials with standard shoes and two 30s trials with orthopaedic shoes. A triaxial motion sensor recorded 3D accelerations at the lower back level. LDS was assessed by computing divergence exponents in the acceleration signals (maximal Lyapunov exponents). Pain was evaluated with Visual Analogue Scale (VAS). LDS and pain differences between the trials with standard shoes and the trials with orthopaedic shoes were assessed. RESULTS: Orthopaedic shoes significantly improved LDS in the three axes (medio-lateral: 10% relative change, paired t-test p < 0.001; vertical: 9%, p = 0.03; antero-posterior: 7%, p = 0.04). A significant decrease in pain level (VAS score -29%) was observed. CONCLUSIONS: Footwear adaptation led to pain relief and to improved foot & ankle proprioception. It is likely that that enhancement allows patients to better control foot placement. As a result, higher dynamic stability has been observed. LDS seems therefore a valuable index that could be used in early evaluation of footwear outcome in clinical settings.