Partial Least Squares Discriminant Analysis Method Research Articles

Classification of healthy and cancerous colon cells by Fourier transform infrared spectroscopy

Colorectal cancer is one of the most diagnosed types of cancer in developed countries. Current diagnostic methods are partly dependent on pathologist experience and laboratories instrumentation. In this study, we used Fourier Transform Infrared (FTIR) spectroscopy in transflection mode, combined with Principal Components Analysis followed by Linear Discriminant Analysis (PCA-LDA) and Partial Least Squares – Discriminant Analysis (PLS-DA), to build a classification algorithm to diagnose colon cancer in cell samples, based on absorption spectra measured in two spectral ranges of the mid-infrared spectrum. In particular, PCA technique highlights small biochemical differences between healthy and cancerous cells: these are related to the larger lipid content in the former compared with the latter and to the larger relative amount of protein and nucleic acid components in the cancerous cells compared with the healthy ones. Comparison of the classification accuracy of PCA-LDA and PLS-DA methods applied to FTIR spectra measured in the 1000–1800 cm−1 (low wavenumber range, LWR) and 2700–3700 cm−1 (high wavenumber range, HWR) remarks that both algorithms are able to classify hidden class FTIR spectra with excellent accuracy (100 %) in both spectral regions. This is a hopeful result for clinical translation of infrared spectroscopy: in fact, it makes reliable the predictions obtained using FTIR measurements carried out only in the HWR, in which the glass slides used in clinical laboratories are transparent to IR radiation.

Deposit type discrimination based on trace elements in sphalerite

Sphalerite can form in different geological environments and is the main ore mineral of most Pb-Zn deposits. Trace elements in sphalerite combined with machine learning (ML) methods have been used for deposit type discrimination. However, the performance of different multivariate and ML methods is not systematically compared. Here more than four thousand laser ablation inductively coupled plasma mass spectrometry data from more than one hundred deposits were compiled and investigated by different multivariate and ML methods. Unsupervised multivariate and ML methods including principal component analysis (PCA) and t-distributed stochastic neighbor embedding (t-SNE), and supervised multivariate and ML methods including partial least squares-discriminant analysis (PLS-DA), logistic regression (LR), support vector machine (SVM), K-nearest neighbors (KNN), random forest (RF), naïve Bayes (NB), and neural network (NN) are used to classify deposit types. The models are constructed for five deposit types, Mississippi Valley-type (MVT), Volcanogenic Massive Sulfide (VMS), sedimentary exhalative (SEDEX), Epithermal, and Skarn deposits, based on twelve elements in sphalerite, Fe, Pb, Mn, Co, Cu, Ga, Ge, Ag, Cd, In, Sn, and Sb.The PCA and PLS-DA methods show similar clustering results, where Epithermal, MVT, SEDEX, and Skarn deposits can be separated from other deposit types. The t-SNE method is only efficient in discriminating MVT, SEDEX, and Epithermal deposits. The KNN, RF, and NN methods show better performance than LR, SVM, and NB in terms of average accuracy and precision. The PCA, t-SNE, and PLS-DA methods have advantages in data visualization, whereas KNN, NN, and RF methods are more suitable for classification tasks. Feature importance analyses of the PLS-DA and RF methods show that the most important discriminant elements in the classification are Ge, Mn, In, Sn, Ga, and Sb. The models based on PCA, PLS-DA, KNN, NN, and RF are used to predict sphalerite samples with uncertain origins. All the models give similar classification results for deposit types, which are consistent with the ore-forming geological setting. The multivariate and ML models, including discrimination diagrams of PCA and PLS-DA, can be used as the first screening for the type of Pb-Zn mineralization in mineral exploration.

Multi-ensemble machine learning framework for omics data integration: A case study using breast cancer samples

Integration of voluminous omics data aids to unravel biological complexities associated with different disease phenotypes. Machine learning (ML) approaches provide insightful techniques for systematic multi-omics data integration. In this study, survival prediction of breast cancer patients was undertaken using omics data of 302 female patients from The Cancer Genome Atlas (TCGA). The data included gene expression, miRNA expression, DNA methylation and copy number variation. Three computational multi-ensemble ML pipelines were tested using Support Vector Machine (SVM), Random Forest (RF) and Partial Least Squares-Discriminant Analysis (PLS-DA) algorithms. To overcome the limitations associated with univariate feature selection criteria, the ML pipelines were built along with latent factors obtained by multivariate dimension reduction method. This facilitated investigation of background genetic networks and identification of potential hub genes. Analysis of the results obtained revealed that SVM with PLS-DA method (integrated with gene expression, DNA methylation, and miRNA expression modalities) was the best-performing model with an Area Under Curve (AUC) of 89% and an accuracy of 83% for survival prediction. This study not only corroborated previously reported breast cancer-specific prognostic biomarkers but also predicted additional potential biomarkers. The work demonstrates the effective use of a multi-ensemble ML model with efficient feature selection methods as a robust protocol for cancer genotype to phenotype correlation.

Comparing baseline correction algorithms in discriminating brownish soils from five proximity locations based on UPLC and PLS-DA methods.

UPLC data of soil interfere with baseline drifts.BC can improve the quality of the pixel-level UPLC data.MW emerges as the most desired algorithm in improving the quality of UPLC data of soil.

Label-Free Identification of Biochemical Footprints of Key Mutations in Acute Myeloid Leukemia Using Raman Spectroscopy

Introduction: With increasing understanding of the complexity and heterogeneity of acute myeloid leukemia (AML), the diagnostic process becomes more complicated and time-consuming. Therefore, there is a need to supplement AML diagnostics with a quick method to identify key, druggable genetic lesions that serve as a target for small molecule drugs. To address this need, we used Raman spectroscopy (RS)-based imaging methods. RS uses non-elastic scattering of laser light on cell biomolecules. When coupled to computational image segmentation and analysis, composition of the analyzed fragment of the cytoplasm can be characterized. In this way, by mapping the focal surface of the cell, we obtain a multidimensional image containing information about the distribution of its biochemical components. In our study, we utilized RS microscopy to investigate how single mutations affect the Raman spectral images of leukemic cells. The primary objective was to determine whether this technique could be utilized to identify mutations in key genes related to leukemia development. Methods: Genetic models of AML were generated by introducing wild-type and mutated forms of three frequently mutated and/or druggable genes (FLT3, IDH1/2, and MLL) into myeloid cell lines (THP1, HEL, HL60). The presence of the desired phenotype was confirmed using flow cytometry, immunoblotting, high-throughput kinase activity assay (Pamgene), and confocal microscopy. For RS experiments, the cells were fixed, and subjected to Raman imaging measurements using the WITec Alpha 300 microscope. Reference or control samples consisted of cells with wild-type genes (WT). The recorded hyperspectral data was subjected to multidimensional chemometric analysis, including k-means cluster analysis (KMCA), principal component analysis (PCA) and partial least squares discriminant analysis (PLS-DA). To validate the findings from the cell models, primary AML cells from 30 patients diagnosed with FLT-3 ITD, MLL or IDH1/2 mutations were examined with RS. Peripheral blood mononuclear cells (PBMC) isolated from healthy donors blood served as reference. Results: The spectral differences between cells with wild-type FLT3 and FLT3/ITD were subtle, yet significant. The results revealed distinct variances in lipid (1440, 1240, 1660, and 2850 cm -1) and hemoprotein content (740-760 cm -1). Additionally, we successfully distinguished cells harboring IDH1/2 and MLL mutations from those with the wild-type genes. In these cases Raman bands corresponding to protein and nucleic acids composition were changed in comparison to control cells. Raman features in PCA loadings were also related to proteins and nucleic acids. Using the PLS-DA method and simple AI model, we built successful algorithm for distinguishing transgenic models carrying mutation from wild-type cells. PCA analysis also enabled us to distinguish AML primary cells from PBMC. Analysis weather primary cells carrying the selected mutations can be distinguished from non-mutated AML cells is concordant with our findings on models, however, it requires further analysis in larger group of patients to increase statistical power of these observations. Conclusions: Collectively, these findings suggest that Raman spectroscopy coupled with appropriate chemometric analysis may serve as rapid, label-free tool for identification of key druggable mutations in AML. Acknowledgements: The studies were performed as a part of the “ Label-free and rapid optical imaging, detection and sorting of leukemia cells” project carried out within the Team-Net programme of the Foundation for Polish Science co-financed by the European Union under the European Regional Development Fund.

Illicit Drug Analysis in Blood Samples with Multivariate Analysis Using Surface-Enhanced Raman Spectroscopy

This study aims to discriminate different types of illicit drugs (MDMA and THC) in blood samples using surface-enhanced Raman spectroscopy (SERS) combined with chemometric techniques including principal components analysis (PCA) and partial least squares discriminant analysis (PLS-DA). A PLS-DA classification model was built using a training data set containing Raman spectra from control and experimental groups (drug-detected blood). PLS-DA was performed for discrimination and classification among blood samples. The scores obtained in the PLS-DA model were used to evaluate the performance of the created model. The leave one out cross-validation (LOOCV) method was used for calibration and validation of the PLS-DA model. In the study, it was observed that the SERS method and chemometric techniques together could be used in drug analysis, even at low concentrations in complex body fluids such as blood. As a result, Raman spectroscopy with PCA and PLS-DA methods of data analysis could be used extensively to build similar or different classification models.

The classification performance of multivariate curve resolution-discriminant analysis: A comparative study

Pattern recognition models establish the relationship between the selected variables and objects' belonging to different classes. Variations in measured variables, different object patterns, and relationship models explain the need to develop different algorithms. Partial least squares-discriminant analysis (PLS-DA) is based on the projection of object responses in the space of latent vectors of the PLS calibration model. Recently, Multivariate Curve Resolution-Alternating Least Squares (MCR-ALS) method has been introduced for discriminant analysis by our group. The possibility of incorporation of real physical and chemical information in the MCR decomposition of data leads to solutions that directly have meaning and simplifies the interpretation of results. Considering the new application of MCR in supervised pattern recognition, there is still a need to evaluate and compare the efficiency of MCR-DA with tested methods such as PLS-DA.Detection and discrimination of extra virgin olive oil from four other classes of pure edible oils and also their mixture with extra virgin olive oil, as adulteration was chosen just as a typical system to evaluate and compare the efficiency of MCR-DA and PLS-DA methods. Raman spectroscopy was used forresponse measurement of oil samples. Different classifications were made between extra virgin olive oil and diverse combinations of four other edible oils, and similar sets were used for the comparative evaluation of the two discrimination methods. Various criteria evaluated the performance of the two methods and in most cases the new MCR-DA method was comparable to PLS-DA. However, MCR-DA can assign pure Raman spectra resulting from decomposition to different classes, which can provide direct information about the components of the studied system.

Near infrared spectroscopy (NIR) and chemometrics methods to identification of haploids in maize

Haploids in maize are selected through visual analysis, whether by use of expression of the R1-navajo marker in seeds, or by the phenotype of haploid plants in the field. These methods of identification of haploid plants demand space, resources, and time. Near infrared spectroscopy (NIR), together with multivariate calibration methods, is an alternative for identification of haploid seeds and plants in a quick and accurate way. The aim of the present study was to use NIR to classify haploid and diploid seeds and differentiate haploid, doubled haploid, and diploid plants originating from different genotypes used as donor. In the first experiment, haploid and diploid maize seeds were analyzed from four genotypes used as parents. In another experiment, diploid, haploid, and doubled haploid plants from 34 source populations were evaluated and differentiated. Spectral data were collected from all the seed and leaf samples using the portable MicroNIR device. Then the data were pre-processed by principal component analysis and by partial least squares discriminant analysis (PLS-DA). The use of the portable MicroNIR, together with the PLS-DA method, enabled classification of haploid and diploid seeds and haploid, doubled haploid, and diploid plants with 100% accuracy, constituting a fast, simple, non-destructive, and reliable methodology for identification of haploids in maize, and it is more effective than the visual analysis.

Synchrotron-Infrared Microspectroscopy of Live Leishmania major Infected Macrophages and Isolated Promastigotes and Amastigotes.

The prevalence of neglected tropical diseases (NTDs) is advancing at an alarming rate. The NTD leishmaniasis is now endemic in over 90 tropical and sub-tropical low socioeconomic countries. Current diagnosis for this disease involves serological assessment of infected tissue by either light microscopy, antibody tests, or culturing with in vitro or in vivo animal inoculation. Furthermore, co-infection by other pathogens can make it difficult to accurately determine Leishmania infection with light microscopy. Herein, for the first time, we demonstrate the potential of combining synchrotron Fourier-transform infrared (FTIR) microspectroscopy with powerful discrimination tools, such as partial least squares-discriminant analysis (PLS-DA), support vector machine-discriminant analysis (SVM-DA), and k-nearest neighbors (KNN), to characterize the parasitic forms of Leishmania major both isolated and within infected macrophages. For measurements performed on functional infected and uninfected macrophages in physiological solutions, the sensitivities from PLS-DA, SVM-DA, and KNN classification methods were found to be 0.923, 0.981, and 0.989, while the specificities were 0.897, 1.00, and 0.975, respectively. Cross-validated PLS-DA models on live amastigotes and promastigotes showed a sensitivity and specificity of 0.98 in the lipid region, while a specificity and sensitivity of 1.00 was achieved in the fingerprint region. The study demonstrates the potential of the FTIR technique to identify unique diagnostic bands and utilize them to generate machine learning models to predict Leishmania infection. For the first time, we examine the potential of infrared spectroscopy to study the molecular structure of parasitic forms in their native aqueous functional state, laying the groundwork for future clinical studies using more portable devices.

Evaluation of Dynamic Changes and Regularity of Volatile Flavor Compounds for Different Green Plum (Prunus mume Sieb. et Zucc) Varieties during the Ripening Process by HS-GC-IMS with PLS-DA.

Headspace gas chromatography-ion mobility spectrometry and partial-least-squares discriminant analysis (PLS-DA) were adopted to analyze the rule of change in flavor substances for different varieties of green plums at different levels of maturity (S1-immature, S2-commercially mature, and S3-fully mature). The results showed that 68 kinds of volatile flavor substances were identified in all green plum samples. The types and contents of such volatile substances experienced a V-shaped trend with an increasing degree of green plum maturity. During the S1 and S2 stages, aldehydes, ketones, and a small amount of alcohols were the main volatile flavor substances in the green plum samples. During the S3 stage, esters and alcohols were the most important volatile flavor components in the green plum pulp samples, followed by terpenes and ketones. YS had the most types and highest contents of volatile flavor substances in three stages, followed by GC and DZ. By using the PLS-DA method, this study revealed the differences in flavor of the different varieties of green plums at different maturity stages, and it identified eight common characteristic volatile flavor substances, such as ethyl acetate, 3-methylbutan-1-ol, and 2-propanone, produced by the different green plum samples during the ripening process, as well as the characteristic flavor substances of green plums at each maturity stage (S1-S3).

Bayesian Fusion Model Enhanced Codfish Classification Using Near Infrared and Raman Spectrum.

In this study, a Bayesian-based decision fusion technique was developed for the first time to quickly and non-destructively identify codfish using near infrared (NIRS) and Raman spectroscopy (RS). NIRS and RS spectra from 320 codfish samples were collected, and separate partial least squares discriminant analysis (PLS-DA) models were developed to establish the relationship between the raw data and cod identity for each spectral technique. Three decision fusion methods: decision fusion, data layer or feature layer, were tested and compared. The decision fusion model based on the Bayesian algorithm (NIRS-RS-B) was developed on the optimal discrimination features of NIRS and RS data (NIRS-RS) extracted by the PLS-DA method whereas the other fusion models followed conventional, non-Bayesian approaches. The Bayesian model showed enhanced classification metrics (92% sensitivity, 98% specificity, 98% accuracy) that were significantly superior to those demonstrated by any of other two spectroscopic methods (NIRS, RS) and the two data fusion methods (data layer fused, NIRS-RS-D, or feature layer fused, NIRS-RS-F). This novel proposed approach can provide an alternative classification for codfish and potentially other food speciation cases.

Reliable classification of nutmeg fruits from some geographical origins using SWNIR spectroscopy associated with PLS-DA and SVM methods: Case Study in Aceh Province, Indonesia

Essential oils of nutmeg (Myristica fragrans) can have different qualities due to the soil and industrial process, as Indonesia’s main nutmeg production areas are enormous, especially in Aceh province. The lack of information on detecting chemical constituents and classifying nutmeg from Aceh province makes the price of nutmeg from Aceh unable to compete. Therefore, this study aims to reliably classify nutmeg fruits of some geographical origins in Aceh Province using shortwave near-infrared (SWNIR) spectroscopy associated with partial least squares discriminant analysis (PLS-DA) and support vector machine (SVM). The SWNIR used in this study has a wavelength of 381 to 1065 nm at a resolution of 3 nm and a diffuse reflectance mode. The fruit comes from four geographical areas of origin in the province of Aceh. A total of forty SWNIR spectral data were further classified using two algorithms, namely PLS-DA and SVM. The results show that SWNIR can correctly classify nutmeg from some sub-districts in Aceh province with the help of the PLS-DA and SVM algorithms. From this study, the following SVM algorithm is recommended to be used to classify nutmeg from the province of Aceh with an outstanding level of precision.

A comparative study on classification of edible vegetable oils by infrared, near infrared and fluorescence spectroscopy combined with chemometrics

Driven by economic benefits like any other foods, vegetable oil has long been plagued by mislabeling and adulteration. Many studies have addressed the field of classification and identification of vegetable oils by various analysis techniques, especially spectral analysis. A comparative study was performed using Fourier transform infrared spectroscopy (FTIR), visible near-infrared spectroscopy (Vis-NIR) and excitation-emission matrix fluorescence spectroscopy (EEMs) combined with chemometrics to distinguish different types of edible vegetable oils. FTIR, Vis-NIR and EEMs datasets of 147 samples of five vegetable oils from different brands were analyzed. Two types of pattern recognition methods, principal component analysis (PCA)/multi-way principal component analysis (M-PCA) and partial least squares discriminant analysis (PLS-DA)/multilinear partial least squares discriminant analysis (N-PLS-DA), were used to resolve these data and distinguish vegetable oil types, respectively. PCA/M-PCA analysis exhibited that three spectral data of five vegetable oils showed a clustering trend. The total correct recognition rate of the training set and prediction set of FTIR spectra of vegetable oil based on PLS-DA method are 100%. The total recognition rate of Vis-NIR based on PLS-DA are 100% and 97.96%. However, the total correct recognition rate of training set and prediction set of EEMs data based on N-PLS-DA method is 69.39% and 75.51%, respectively. The comparative study showed that FTIR and Vis-NIR combined with chemometrics were more suitable for vegetable oil species identification than EEMs technique. The reason may be concluded that almost all chemical components in vegetable oil can produce FTIR and NIR absorption, while only a small amount of fluorophores can produce fluorescence. That is, FTIR and NIR can provide more spectral information than EEMs. Analysis of EEMs data using self-weighted alternating trilinear decomposition (SWATLD) also showed that fluorophores were a few and irregularly distributed in vegetable oils.

Multivariate statistical approach and machine learning for the evaluation of biogeographical ancestry inference in the forensic field

The biogeographical ancestry (BGA) of a trace or a person/skeleton refers to the component of ethnicity, constituted of biological and cultural elements, that is biologically determined. Nowadays, many individuals are interested in exploring their genealogy, and the capability to distinguish biogeographic information about population groups and subgroups via DNA analysis plays an essential role in several fields such as in forensics. In fact, for investigative and intelligence purposes, it is beneficial to inference the biogeographical origins of perpetrators of crimes or victims of unsolved cold cases when no reference profile from perpetrators or database hits for comparative purposes are available. Current approaches for biogeographical ancestry estimation using SNPs data are usually based on PCA and Structure software. The present study provides an alternative method that involves multivariate data analysis and machine learning strategies to evaluate BGA discriminating power of unknown samples using different commercial panels. Starting from 1000 Genomes project, Simons Genome Diversity Project and Human Genome Diversity Project datasets involving African, American, Asian, European and Oceania individuals, and moving towards further and more geographically restricted populations, powerful multivariate techniques such as Partial Least Squares-Discriminant Analysis (PLS-DA) and machine learning techniques such as XGBoost were employed, and their discriminating power was compared. PLS-DA method provided more robust classifications than XGBoost method, showing that the adopted approach might be an interesting tool for forensic experts to infer BGA information from the DNA profile of unknown individuals, but also highlighting that the commercial forensic panels could be inadequate to discriminate populations at intra-continental level.

Effects of different thermal processing methods on nutrients and flavor of Toona sinensis

The nutrients, bioactives, and flavor compounds of Toona sinensis (TS) processed by blanching, steaming, roasting, and frying were analyzed. It was found that the loss of nutrients, bioactives, and flavor compounds was highest for fried TS and lowest for steamed TS. The thermal processing method had a significant impact on the quality characteristics of processed TS. Eleven new compounds were identified in raw and processed TS, while 2-mercapto-3,4-dimethyl-2,3-dihydrothiophene had the highest odor activity value (OAV) among the volatiles in all the samples. The principal component analysis (PCA) results indicated that the volatile compounds profiles in different processed TS were completely different. Eight compounds identified by the partial least square discriminant analysis (PLS-DA) method were the markers to distinguish the different processed TS. Steaming was the most suitable thermal processing method for TS. This study provides the theoretical basis for the selection of optimized processing methods for TS. Novelty Impact Statement The nutrients, bioactives, and flavor compounds of Toona sinensis processed by steaming, boiling, frying, and roasting were analyzed, and the thermal processing method had a significant impact on the quality characteristics of processed Toona sinensis. Eleven new compounds were identified in Toona sinensis; GC-IMS analysis offered their distinct advantages to the aroma analysis of Toona sinensis. Eight compounds identified by the PLS-DA method were the markers to distinguish the Toona sinensis. Steaming was the most suitable thermal processing method for Toona sinensis.

Forensic analysis of cigarette ash using ATR-FTIR spectroscopy and chemometric methods

Cigarette ash is frequently found in crime scenes involving murder, suicide, rape, dacoity, or assault cases, but it is frequently overlooked as potential evidence by investigators. This ash as evidence can assist an investigator in figuring out the number of people who were there at the time of the crime as well as narrowing down the suspect pool. Because any type of ash, such as burnt paper, dhoop, agarbatti, cloth, tandoor, and so on, could be present at the crime scene, the first objective of this research study proposed a method for the characterization and discrimination of cigarette ash from other types using ATR-FTIR Spectroscopy in conjunction with statistical models PCA and PLSDA. In the second objective, this approach was used for the discrimination of cigarette and bidi ash. Finally, the same strategy is used to distinguish between ashes from three different cigarette brands because some people prefer one brand over another, which may be useful in certain situations. For this research study, we collected a total of 94 samples (47 unsmoked/fresh cigarette sticks of 20 brands from local vendors and 47 samples as other ash types from the surrounding area), 12 samples of bidi/hand-rolled cigarettes from different brands for cigarette ash discrimination, and 30 samples (10 sticks for a single cigarette brand) for brand discrimination. After smouldering the collected sample sticks in the laboratory, ash spectra were obtained for all of them for cigarette ash discrimination and brand determination. Differences in peak patterns, their absorbance, and the presence or absence of specific components in the fingerprint region were investigated for discriminatory purposes. Because FTIR spectra evaluation is time-consuming and prone to human error, the data was further analysed using multivariate techniques such as PCA (principal component analysis) and PLS-DA (Partial Least Square Discriminant Analysis) to improve sample discrimination. In the PCA plots, there was a degree of sample segregation. Therefore, for classifying the data into groups and predicting their outcomes, the PLS-DA method was employed, and a satisfactory categorization of samples was accomplished. The performance of the PLS-DA model is further validated using blind testing of unknown samples. As a result of the current research work, a non-destructive, efficient, and non-invasive approach for analysing cigarette ash has been established, which would aid in crime scene investigation.

A Joint Bayesian Framework Based on Partial Least Squares Discriminant Analysis for Finger Vein Recognition

Finger vein recognition has attracted considerable attention from the biometric identification technology community owing to its convenience and security. Unlike most previous works only pay attention to one part of finger vein recognition, we propose a joint Bayesian framework in this paper, which is based on partial least squares discriminant analysis (PLS-DA). It involves three major stages: 1) robust feature description, 2) discriminative feature mapping, and 3) separative verification. In stage 1), we extract line responses and orientation of finger veins using a bank of Gabor filter, and histograms are constructed in local patches as primitive features. Subsequently, in stage 2), a discriminant feature mapping based the PLS-DA (PLS-DA-FM) method is proposed to project these primitive features into low-dimensional forms in a supervised manner. Thus, highly compact and discriminative features are obtained in this stage. Finally, in stage 3), we directly build a Bayesian model based on the joint distribution of finger vein feature pairs to measure the similarity between the features. Extensive experiments on five finger vein datasets demonstrate the superior performance of the proposed method to most state-of-the-art finger vein recognition methods.

Detection of adulterations in a valuable Brazilian honey by using spectrofluorimetry and multiway classification

Spectrofluorimetry combined with multiway chemometric tools were applied to discriminate pure Aroeira honey samples from samples adulterated with corn syrup, sugar cane molasses and polyfloral honey. Excitation emission spectra were acquired for 232 honey samples by recording excitation from 250 to 500 nm and emission from 270 to 640 nm. Parallel factor analysis (PARAFAC), partial least squares discriminant analysis (PLS-DA), unfolded PLS-DA (UPLS-DA) and multilinear PLS-DA (NPLS-DA) methods were used to decompose the spectral data and build classification models. PLS-DA models presented poor classification rates, demonstrating the limitation of the traditional two-way methods for this dataset, and leading to the development of three-way classification models. Overall, UPLS-DA provided the best classification results with misclassification rates of 4% and 8% for the training and test sets, respectively. These results showed the potential of the proposed method for routine laboratory analysis as a simple, reliable, and affordable tool.

Analysis of Various Techniques for Solving the Problem of Big Data Classification

The purpose of this study is to assess the effectiveness of various algorithms for big data classification, namely, partial least squares discriminant analysis (PLS-DA), NaiveBayes (NBC) and K-Nearest Neighbor (KNN) based on the Hadoop MapReduce approach. The effectiveness of the approaches is compared to the classification of big data sets of average shot lengths (CSV). It has been shown that in accordance with the data set size, the PLS-DA classification accuracy increases and reaches 82%, and the computation time goes up to 45 seconds. The analysis of various classifiers showed that high accuracy rates for the PLS-DA classifier are ensured by a high percentage of positive and negative cases properly classified, and lower accuracy for KNN and NaiveBayes is justified by a high percentage of false-positive and false-negative indicators. It is concluded that the optimal classifier is the PLS-DA method, which allows one to classify a large amount data with high accuracy in a short time.

Development of a genomic tool for breed assignment by comparison of different classification models: Application to three local cattle breeds.

Assignment of individual cattle to a specific breed can often not rely on pedigree information. This is especially the case for local breeds for which the development of genomic assignment tools is required to allow individuals of unknown origin to be included to their herd books. A breed assignment model can be based on two specific stages: (a) the selection of breed-informative markers and (b) the assignment of individuals to a breed with a classification method. However, the performance of combination of methods used in these two stages has been rarely studied until now. In this study, the combination of 16 different SNP panels with four classification methods was developed on 562 reference genotypes from 12 cattle breeds. Based on their performances, best models were validated on three local breeds of interest. In cross-validation, 14 models had a global cross-validation accuracy higher than 90%, with a maximum of 98.22%. In validation, best models used 7,153 or 2,005 SNPs, based on a partial least squares-discriminant analysis (PLS-DA) and assigned individuals to breeds based on nearest shrunken centroids. The average validation sensitivity of the first two best models for the three local breeds of interest were 98.33% and 97.5%. Moreover, results reported in this study suggest that further studies should consider the PLS-DA method when selecting breed-informative SNPs.