Modern statistical models for forensic fingerprint examinations : a critical review

Over the last decade, the development of statistical models in support of forensic fingerprint identification has been the subject of increasing research attention, spurned on recently by commentators who claim that the scientific basis for fingerprint identification has not been adequately demonstrated. Such models are increasingly seen as useful tools in support of the fingerprint identification process within or in addition to the ACE-V framework. This paper provides a critical review of recent statistical models from both a practical and theoretical perspective. This includes analysis of models of two different methodologies: Probability of Random Correspondence (PRC) models that focus on calculating probabilities of the occurrence of fingerprint configurations for a given population, and Likelihood Ratio (LR) models which use analysis of corresponding features of fingerprints to derive a likelihood value representing the evidential weighting for a potential source.

A unified approach for representing rows and columns in contingency tables

By using suitable parameters, we present a uni¯ed aproach for describing four methods for representing categorical data in a contingency table. These methods include:correspondence analysis (CA), the alternative approach using Hellinger distance (HD),the log-ratio (LR) alternative, which is appropriate for compositional data, and theso-called non-symmetrical correspondence analysis (NSCA). We then make an appropriate comparison among these four methods and some illustrative examples are given.Some approaches based on cumulative frequencies are also linked and studied usingmatrices.Key words: Correspondence analysis, Hellinger distance, Non-symmetrical correspondence analysis, log-ratio analysis, Taguchi inertia

Genetics of leprosy reactions: an overview

Type-1 (T1R) and Type-2 (T2R) leprosy reactions (LR), which affect up to 50% of leprosy patients, are aggressive inflammatory episodes of sudden onset and highly variable incidence across populations. LR are often diagnosed concurrently with leprosy, but more frequently occur several months after treatment onset. It is not uncommon for leprosy patients to develop recurring reactional episodes; however, they rarely undergo both types of LR. Today, LR are the main cause of permanent disabilities associated with leprosy and represent a major challenge in the clinical management of leprosy patients. Although progress has been made in understanding the immunopathology of LR, the factors that cause a leprosy patient to suffer from LR are largely unknown. Given the impact that ethnic background has on the risk of developing LR, host genetic factors have long been suspected of contributing to LR. Indeed, polymorphisms in seven genes [Toll-like receptors (TLR)1, TLR2, nucleotide-binding oligomerisation domain containing 2, vitamin D receptor, natural resistance-associated macrophage protein 1, C4B and interleukin-6] have been found to be associated with one or more LR outcomes. The identification of host genetic markers with predictive value for LR would have a major impact on nerve damage control in leprosy. In this review, we present the recent advances achieved through genetic studies of LR.

Good agreements between self and clinician-collected specimens for the detection of human papillomavirus in Brazilian patients

Women infected with human papillomavirus (HPV) are at a higher risk of developing cervical lesions. In the current study, self and clinician-collected vaginal and cervical samples from women were processed to detect HPV DNA using polymerase chain reaction (PCR) with PGMY09/11 primers. HPV genotypes were determined using type-specific PCR. HPV DNA detection showed good concordance between self and clinician-collected samples (84.6%; kappa = 0.72). HPV infection was found in 30% women and genotyping was more concordant among high-risk HPV (HR-HPV) than low-risk HPV (HR-HPV). HPV16 was the most frequently detected among the HR-HPV types. LR-HPV was detected at a higher frequency in self-collected; however, HR-HPV types were more frequently identified in clinician-collected samples than in self-collected samples. HPV infections of multiple types were detected in 20.5% of clinician-collected samples and 15.5% of self-collected samples. In this study, we demonstrated that the HPV DNA detection rate in self-collected samples has good agreement with that of clinician-collected samples. Self-collected sampling, as a primary prevention strategy in countries with few resources, could be effective for identifying cases of HR-HPV, being more acceptable. The use of this method would enhance the coverage of screening programs for cervical cancer.

Antiretroviral therapy as a factor protective against anal dysplasia in HIV-infected males who have sex with males.

OBJECTIVES Chronic infection with oncogenic HPV genotype is associated with the development of anal dysplasia. Antiretroviral therapy (ART) has been shown to decrease the incidence of cervical carcinoma in women with HIV. We sought to: 1) describe the prevalence and grade of anal dysplasia and HPV infection in our study subjects; 2) analyze the grade of correlation between anal cytology, PCR of high-risk HPV, and histology; 3) identify the factors associated with the appearance of ≥ AIN2 lesions. DESIGN Cross-sectional, prospective study. METHODS A cohort of HIV-positive males (n = 140, mean age  = 37 years) who have sex with males (MSM) had epidemiological, clinical and analytical data collected. Anal mucosa samples were taken for cytology, HPV PCR genotyping, and anoscopy for histological analysis. RESULTS Within the cohort, 77.1% were being treated with ART, 8.5% anoscopy findings were AIN2, and 11.4% carcinoma in situ; 74.2% had high-risk (HR), 59.7% low-risk (LR) HPV genotypes and 46.8% had both. The combination of cytology with PCR identifying HR-HPV better predicts the histology findings than either of these factors alone. Logistic regression highlighted ART as a protective factor against ≥ AIN2 lesions (OR: 0.214; 95%CI: 0.054-0.84). Anal/genital condylomas (OR: 4.26; 95%CI: 1.27-14.3), and HPV68 genotype (OR: 10.6; 95%CI: 1.23-91.47) were identified as risk factors. CONCLUSIONS In our cohort, ART has a protective effect against dysplastic anal lesions. Anal/genital warts and HPV68 genotype are predictors of ≥ AIN2 lesions. Introducing PCR HPV genotype evaluation improves screening success over that of cytology alone.

REST Regulates Oncogenic Properties of Glioblastoma Stem Cells.

Glioblastoma multiforme (GBM) tumors are the most common malignant primary brain tumors in adults. Although many GBM tumors are believed to be caused by self-renewing, glioblastoma-derived stem-like cells (GSCs), the mechanisms that regulate self-renewal and other oncogenic properties of GSCs are only now being unraveled. Here we showed that GSCs derived from GBM patient specimens express varying levels of the transcriptional repressor repressor element 1 silencing transcription factor (REST), suggesting heterogeneity across different GSC lines. Loss- and gain-of-function experiments indicated that REST maintains self-renewal of GSCs. High REST-expressing GSCs (HR-GSCs) produced tumors histopathologically distinct from those generated by low REST-expressing GSCs (LR-GSCs) in orthotopic mouse brain tumor models. Knockdown of REST in HR-GSCs resulted in increased survival in GSC-transplanted mice and produced tumors with higher apoptotic and lower invasive properties. Conversely, forced expression of exogenous REST in LR-GSCs produced decreased survival in mice and produced tumors with lower apoptotic and higher invasive properties, similar to HR-GSCs. Thus, based on our results, we propose that a novel function of REST is to maintain self-renewal and other oncogenic properties of GSCs and that REST can play a major role in mediating tumorigenicity in GBM. STEM CELLS 2012;30:405-414.

Diagnostic Accuracy of (18)F-FDG-PET and PET/CT in the Differential Diagnosis between Malignant and Benign Pleural Lesions: A Systematic Review and Meta-Analysis.

RATIONALE AND OBJECTIVES: To systematically review and meta-analyze published data about the diagnostic accuracy of fluorine-18-fluorodeoxyglucose ((18)F-FDG) positron emission tomography (PET) and PET/computed tomography (CT) in the differential diagnosis between malignant and benign pleural lesions. METHODS AND MATERIALS: A comprehensive literature search of studies published through June 2013 regarding the diagnostic performance of (18)F-FDG-PET and PET/CT in the differential diagnosis of pleural lesions was carried out. All retrieved studies were reviewed and qualitatively analyzed. Pooled sensitivity, specificity, positive and negative likelihood ratio (LR+ and LR-) and diagnostic odds ratio (DOR) of (18)F-FDG-PET or PET/CT in the differential diagnosis of pleural lesions on a per-patient-based analysis were calculated. The area under the summary receiver operating characteristic curve (AUC) was calculated to measure the accuracy of these methods. Subanalyses considering device used (PET or PET/CT) were performed. RESULTS: Sixteen studies including 745 patients were included in the systematic review. The meta-analysis of 11 selected studies provided the following results: sensitivity 95% (95% confidence interval [95%CI]: 92-97%), specificity 82% (95%CI: 76-88%), LR+ 5.3 (95%CI: 2.4-11.8), LR- 0.09 (95%CI: 0.05-0.14), DOR 74 (95%CI: 34-161). The AUC was 0.95. No significant improvement of the diagnostic accuracy considering PET/CT studies only was found. CONCLUSIONS: (18)F-FDG-PET and PET/CT demonstrated to be accurate diagnostic imaging methods in the differential diagnosis between malignant and benign pleural lesions; nevertheless, possible sources of false-negative and false-positive results should be kept in mind.

Forensic identification of urine samples: a comparison between nuclear and mitochondrial DNA markers.

Urine samples from 20 male volunteers of European Caucasian origin were stored at 4 degrees C over a 4-month period in order to compare the identification potential of nuclear DNA (nDNA) and mitochondrial DNA (mtDNA) markers. The amount of nDNA recovered from urines dramatically declined over time. Consequently, nDNA likelihood ratios (LRs) greater than 1,000 were obtained for 100, 70 and 55% of the urines analysed after 6, 60 and 120 days, respectively. For the mtDNA, HVI and HVII sequences were obtained for all samples tested, whatever the period considered. Nevertheless, the highest mtDNA LR of 435 was relatively low compared to its nDNA equivalent. Indeed, LRs obtained with only three nDNA loci could easily exceed this value and are quite easier to obtain. Overall, the joint use of nDNA and mtDNA markers enabled the 20 urine samples to be identified, even after the 4-month period.

Biological markers of alcohol consumption in alcoholised drivers: Comparison of capillary electrophoresis (CZE CDT) and a direct immunoassay (N Latex CDT) with the traditional method of anion-exchange chromatography-immunoturbidimetry (%CDT TIA).

Objectives: The aim of this study was to compare specificity and sensitivity of different biological markers that can be used in a forensic field to identify potentially dangerous drivers because of their alcohol habits. Methods: We studied 280 Swiss drivers after driving while under the alcohol influence. 33 were excluded for not having CDT N results, 247 were included (218 men (88%) and 29 women (12%). Mean age was 42,4 (SD:12, min: 20 max: 76). The evaluation of the alcohol consumption concerned the month before the CDT test and was considered as such after the interview: Heavy drinkers (>3 drinks per day): 60 (32.7%), < 3 drinks per day and moderate: 127 (51.4%) 114 (46.5%), abstinent: 60 (24.3%) 51 (21%). Alcohol intake was monitored by structured interviews, self-reported drinking habits and the C-Audit questionnaire as well as information provided by their family and general practitioner. Consumption was quantified in terms of standard drinks, which contain approximately 10 grams of pure alcohol (Ref. WHO). Results: comparison between moderate (less or equal to 3 drinks per day) and excessive drinkers (more than 3 drinks) Marker ROC area 95% CI cut-off sensitivity specificity CDT TIA 0.852 0.786-0917 2.6* 0.93 LR+1.43 0.35 LR-0.192 CDT N latex 0.875 0.821-0.930 2.5* 0.66 LR+ 6.93 0.90 LR- 0.369 Asialo+disialo-tf 0.881 0.826-0.936 1.2* 0.78 LR+4.07 0.80 LR-0.268 1.7° 0.66 LR+8.9 0.93 LR-0.360 GGT 0.659 0.580-0.737 85* 0.37 LR+2.14 0.83 LR-0.764 * cut-off point suggested by the manufacturer ° cut-off point suggested by our laboratory Conclusion: With the cut-off point established by the manufacturer, CDT TIA performed poorly in term of specificity. N latex CDT and CZE CDT were better, especially if a 1.7 cut-off is used with CZE

On bounds for network revenue management

The Network Revenue Management problem can be formulated as a stochastic dynamic programming problem (DP or the\optimal" solution V *) whose exact solution is computationally intractable. Consequently, a number of heuristics have been proposed in the literature, the most popular of which are the deterministic linear programming (DLP) model, and a simulation based method, the randomized linear programming (RLP) model. Both methods give upper bounds on the optimal solution value (DLP and PHLP respectively). These bounds are used to provide control values that can be used in practice to make accept/deny decisions for booking requests. Recently Adelman [1] and Topaloglu [18] have proposed alternate upper bounds, the affine relaxation (AR) bound and the Lagrangian relaxation (LR) bound respectively, and showed that their bounds are tighter than the DLP bound. Tight bounds are of great interest as it appears from empirical studies and practical experience that models that give tighter bounds also lead to better controls (better in the sense that they lead to more revenue). In this paper we give tightened versions of three bounds, calling themsAR (strong Affine Relaxation), sLR (strong Lagrangian Relaxation) and sPHLP (strong Perfect Hindsight LP), and show relations between them. Speciffically, we show that the sPHLP bound is tighter than sLR bound and sAR bound is tighter than the LR bound. The techniques for deriving the sLR and sPHLP bounds can potentially be applied to other instances of weakly-coupled dynamic programming.

Validation procedures in radiological diagnostic models. Neural network and logistic regression

The objective of this paper is to compare the performance of twopredictive radiological models, logistic regression (LR) and neural network (NN), with five different resampling methods. One hundred and sixty-seven patients with proven calvarial lesions as the only known disease were enrolled. Clinical and CT data were used for LR and NN models. Both models were developed with cross validation, leave-one-out and three different bootstrap algorithms. The final results of each model were compared with error rate and the area under receiver operating characteristic curves (Az). The neural network obtained statistically higher Az than LR with cross validation. The remaining resampling validation methods did not reveal statistically significant differences between LR and NN rules. The neural network classifier performs better than the one based on logistic regression. This advantage is well detected by three-fold cross-validation, but remains unnoticed when leave-one-out or bootstrap algorithms are used.

Peletização do lodo de esgoto e adição de CaCO3 na produção de matéria seca e absorção de Zn, Cu e Ni pelo milho em três latossolos

Em experimento de casa de vegetação, realizado em Campinas (SP), no período de janeiro a maio de 1980, estudaram-se os efeitos da peletização do lodo de esgoto, nas doses 0, 1 e 5% (v/v) do material seco, na produção de matéria seca e na absorção de Zn, Cu e Ni pela parte aérea do milho (Zea mays L.), em latossolo roxo (LR), latossolo vermelho-escuro (LE) e latossolo vermelho-amarelo (LV), que receberam ou não adição de CaCO3 suficiente para elevação do pH em água para 6,0. O experimento foi realizado em vasos com dois litros de capacidade, delineados em blocos ao acaso, com três repetições, e os tratamentos arranjados num esquema fatorial. Após 200 dias de incubação dos solos com o lodo e CaCO3, cultivaram-se quatro plantas de milho em cada vaso. A parte aérea foi quantificada 56 dias após a germinação, sendo cortada, seca, pesada e analisada para os elementos Zn, Cu e Ni. Amostras de terra de cada tratamento foram retiradas e tiveram os metais extraídos pelo DTPA. A peletização do lodo de esgoto resultou em diminuição significativa na produção de matéria seca pela parte aérea do milho, nos três solos estudados, em comparação com o lodo não peletizado. A adição de CaCO3 proporcionou aumento significativo na produção de matéria seca do milho apenas nos dois solos mais ácidos (LE e LV) e diminuiu o acúmulo de Zn na parte aérea desse vegetal cultivado no LE, sem distinção quanto ao tipo de lodo aplicado, não se observando diferenças significativas na absorção de Ni, em todos os solos analisados. A incorporação do lodo peletizado , que continha 19% menos Zn e Ni que o não peletizado, resultou em menor absorção de Zn, Cu e Ni nos três solos, exceto para o Ni no LR. Para todos os solos investigados, o método do DTPA mostrou-se mais adequado em prognosticar as quantidades disponíveis de Zn e de Cu para o milho, independentemente do tipo de lodo de esgoto aplicado e da elevação ou não do pH do solo pela adição de CaCO3.

Caracterização de substâncias húmicas extraídas de solos e de lodo orgânico

No presente trabalho, foi estudada a extração de substâncias húmicas (SH) de horizontes A de dois solos do Rio Grande do Sul (Latossolo Roxo (LR), e Brunizém vértico (B)) e de um lodo orgânico (LO), oriundo da fabricação de celulose, empregando-se dois métodos: (1) as SH foram extraídas com solução de NaOH 0,5 mol L-1, após tratamento prévio com HCl 0,5 mol L-1; (2) foi introduzida a extração com solução de pirofosfato de sódio 0,15 mol L-1 (pH 7) antes da etapa alcalina. Em cada etapa, os teores de SH, de ácidos fúlvicos (AF) e de ácidos húmicos (AH) foram determinados por métodos espectroscópicos e gravimétricos e, no método 2, os teores de Fe, Al, Si e Ca foram medidos nos extratos por absorção atômica. A composição elementar de ácidos húmicos foi determinada, e sua estrutura e características químicas foram avaliadas a partir de razões C/N, (atômicas) H/C e O/C e a partir de dados de espectroscopia Ultravioleta/Visível (UV/Vis). No LR, o método 1 extraiu maior quantidade de SH, enquanto, no método 2, a etapa com pirofosfato extraiu apenas AF. Em contrapartida, no solo B, o método 2 foi mais eficiente, verificando-se maior rendimento na etapa com pirofosfato. A partir desses resultados, infere-se que as SH, no LR, interagem principalmente por meio de reação de trocas de ligantes e de pontes de hidrogênio e, no solo B, a interação por meio de pontes de cátions é um mecanismo importante. Os AH extraídos com NaOH apresentaram valores mais elevados para a razão H/C e valores inferiores para a razão O/C do que os extraídos com pirofosfato, indicando que o extrator neutro solubilizou SH de caráter aromático e grau de oxigenação superiores.

Grau de compactação e retenção de água de Latossolos submetidos a diferentes sistemas de manejo

Os estudos do processo de compactação do solo têm envolvido, mais recentemente, alguns ensaios freqüentemente usados na área de mecânica dos solos. Procurando melhor entendimento da aplicação do ensaio de Proctor normal para fins agrícolas, este estudo teve como objetivo avaliar a curva de compactação do solo, obtida por meio deste ensaio, e o efeito do grau de compactação na curva característica de água do solo. No desenvolvimento deste estudo, foram utilizados um Latossolo Roxo (LR), um Latossolo Vermelho-Amarelo (LV) e um Latossolo Vermelho-Escuro (LE), sob três tipos de manejo: cultura anual, pastagem e mata natural, situados no município de Lavras (MG). A umidade ótima de compactação média do LR (0,31 kg kg-1) foi maior do que a do LV (0,22 kg kg-1) e a do LE (0,25 kg kg-1) para as três condições de manejo. Para o LV e para o LE, a densidade do solo máxima aumentou na seguinte ordem: cultura anual, pastagem e mata natural (para o LV, 1,92; 1,94 e 1,95 kg dm-3, respectivamente, e para o LE 1,89; 1,91 e 1,96 kg dm-3, respectivamente); já para o LR, a densidade do solo máxima aumentou na seguinte ordem: mata natural, cultura anual e pastagem (1,83; 1,92 e 1,93 kg dm-3, respectivamente). As curvas características de água do solo foram deslocadas em virtude da variação do grau de compactação. Tais curvas, obtidas para os graus de compactação situados no ramo úmido e no ramo seco, apresentaram tendência de se deslocarem para a direita e para a esquerda, respectivamente, em relação às curvas características de água do solo para o grau de compactação 100%. Para sucções menores do que 10 kPa, 93% das curvas características de água do solo ficaram à direita da curva característica de água para o grau de compactação igual a 100%. Tanto no ramo seco como no ramo úmido, com o aumento do grau de compactação, percebeu-se a diminuição da água disponível às plantas nos três tipos de solo e para todas as condições de manejo.

Compressibilidade de três Latossolos em função da umidade e uso

Por meio do ensaio de compressibilidade, foram estudados os efeitos do manejo e da umidade na pressão de preconsolidação (σp) de três solos: Latossolo Vermelho-Amarelo, Latossolo Roxo e Latossolo Vermelho-Escuro sob cultura anual, mata natural e pastagem, na região de Lavras (MG), em duas profundidades (0-0,03 e 0,27-0,30 m). Para cada condição, foram coletadas cinco amostras indeformadas e uma deformada, com três repetições. As amostras, indeformadas e com diferentes umidades, foram utilizadas no ensaio de compressão uniaxial, obtendo-se as curvas de compressão, das quais foram extraídas as respectivas pressões de preconsolidação. Com as amostras deformadas, determinaram-se os limites de plasticidade e de contração, textura e matéria orgânica. Os modelos de compactação testados foram baseados na pressão de preconsolidação e na umidade do solo. Para uma mesma condição e profundidade, houve diferença significativa entre os valores dos teores de argila e areia nos três solos. Os valores da densidade do solo inicial (Ds i) foram estatisticamente diferentes para todas as condições na mesma profundidade, exceto na camada de 0-0,03 m para a cultura anual. À medida que a umidade do solo aumentou, a pressão de preconsolidação decresceu exponencialmente, indicando uma redução na capacidade de suporte de carga do solo. O LR apresentou, em geral, maior capacidade de suporte de carga do que o LE e LV. Essa maior capacidade de suporte de carga pode estar associada com o seu maior teor de argila e menor teor de areia. A capacidade de suporte de carga na zona de friabilidade variou de 154 a 167 kPa, para o LV; de 77 a 183 kPa, para o LR, e de 77 a 132 kPa, para o LE. As umidades 0,33 a 0,30 kg kg-1, 0,42 a 0,27 kg kg-1, e 0,35 a 0,33 kg kg-1 correspondem à faixa de friabilidade (LP - LC) do LV, LR e LE, respectivamente. O modelo baseado na história de tensão do solo evidenciou o efeito da compactação causada pelas máquinas de preparo do solo na camada de 0,27-0,30 m, para a cultura anual, enquanto, para a pastagem, ocorreu o efeito do pisoteio do gado na camada de 0-0,03 m.