Pub Date : 2022-08-23DOI: 10.3389/frans.2022.961592
Leah D. Pfeifer, M. W. Patabandige, H. Desaire
Applying machine learning strategies to interpret mass spectrometry data has the potential to revolutionize the way in which disease is diagnosed, prognosed, and treated. A persistent and tedious obstacle, however, is relaying mass spectrometry data to the machine learning algorithm. Given the native format and large size of mass spectrometry data files, preprocessing is a critical step. To ameliorate this challenge, we sought to create an easy-to-use, continuous pipeline that runs from data acquisition to the machine learning algorithm. Here, we present a start-to-finish pipeline designed to facilitate supervised and unsupervised classification of mass spectrometry data. The input can be any ESI data set collected by LC-MS or flow injection, and the output is a machine learning ready matrix, in which each row is a feature (an abundance of a particular m/z), and each column is a sample. This workflow provides automated handling of large mass spectrometry data sets for researchers seeking to implement machine learning strategies but who lack expertise in programming/coding to rapidly format the data. We demonstrate how the pipeline can be used on two different mass spectrometry data sets: 1) ESI-MS of fingerprint lipid compositions acquired by direct infusion and, 2) LC-MS of IgG glycopeptides. This workflow is uncomplicated and provides value via its simplicity and effectiveness.
{"title":"Leveraging R (LevR) for fast processing of mass spectrometry data and machine learning: Applications analyzing fingerprints and glycopeptides","authors":"Leah D. Pfeifer, M. W. Patabandige, H. Desaire","doi":"10.3389/frans.2022.961592","DOIUrl":"https://doi.org/10.3389/frans.2022.961592","url":null,"abstract":"Applying machine learning strategies to interpret mass spectrometry data has the potential to revolutionize the way in which disease is diagnosed, prognosed, and treated. A persistent and tedious obstacle, however, is relaying mass spectrometry data to the machine learning algorithm. Given the native format and large size of mass spectrometry data files, preprocessing is a critical step. To ameliorate this challenge, we sought to create an easy-to-use, continuous pipeline that runs from data acquisition to the machine learning algorithm. Here, we present a start-to-finish pipeline designed to facilitate supervised and unsupervised classification of mass spectrometry data. The input can be any ESI data set collected by LC-MS or flow injection, and the output is a machine learning ready matrix, in which each row is a feature (an abundance of a particular m/z), and each column is a sample. This workflow provides automated handling of large mass spectrometry data sets for researchers seeking to implement machine learning strategies but who lack expertise in programming/coding to rapidly format the data. We demonstrate how the pipeline can be used on two different mass spectrometry data sets: 1) ESI-MS of fingerprint lipid compositions acquired by direct infusion and, 2) LC-MS of IgG glycopeptides. This workflow is uncomplicated and provides value via its simplicity and effectiveness.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"41872182","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
The deterioration or oxidation of the mineral oil in transformers poses the risk of short circuits. Convenient and effective methods are expected to be developed. Carbon-based sensor arrays were used in this study to identify the quality variations of mineral oil for oil-filled transformers by odors. The sensitive layers of the odor-sensing system consisted of different types of GC stationary phase materials and carbon black (CB) mixtures. We made a targeted selection of GC materials by utilizing the polarities to make a sensor array based on the distinct components of mineral oil such as alkanes and xylenes by gas chromatography mass spectrometry (GC/MS) analysis. The response characteristics of the sensitive layers were used to recognize the mineral oil odors by machine learning. With laboratory air as the carrier gas, the system could distinguish mineral oil that has been in use for over 20 years from new mineral oil with an accuracy of about 93.8%. The identification accuracy achieved was about 60% for three different concentrations of unused mineral oil and the oxidized mineral oil created by the transformer’s leakage. When detecting the oxidized mineral oil with a concentration of more than 50%, the accuracy rate reached more than 80%. The odor-sensing system in this study will help inspect mineral oils in the transformer and make leakage judgments in a short time.
{"title":"Odor recognition of deteriorated mineral oils using an odor-sensing array","authors":"Yuanchang Liu, Sosuke Akagawa, Rui Yatabe, Takeshi Onodera, Nobuyuki Fujiwara, Hidekazu Takeda, K. Toko","doi":"10.3389/frans.2022.896092","DOIUrl":"https://doi.org/10.3389/frans.2022.896092","url":null,"abstract":"The deterioration or oxidation of the mineral oil in transformers poses the risk of short circuits. Convenient and effective methods are expected to be developed. Carbon-based sensor arrays were used in this study to identify the quality variations of mineral oil for oil-filled transformers by odors. The sensitive layers of the odor-sensing system consisted of different types of GC stationary phase materials and carbon black (CB) mixtures. We made a targeted selection of GC materials by utilizing the polarities to make a sensor array based on the distinct components of mineral oil such as alkanes and xylenes by gas chromatography mass spectrometry (GC/MS) analysis. The response characteristics of the sensitive layers were used to recognize the mineral oil odors by machine learning. With laboratory air as the carrier gas, the system could distinguish mineral oil that has been in use for over 20 years from new mineral oil with an accuracy of about 93.8%. The identification accuracy achieved was about 60% for three different concentrations of unused mineral oil and the oxidized mineral oil created by the transformer’s leakage. When detecting the oxidized mineral oil with a concentration of more than 50%, the accuracy rate reached more than 80%. The odor-sensing system in this study will help inspect mineral oils in the transformer and make leakage judgments in a short time.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-08-17","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45205385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-26DOI: 10.3389/frans.2022.906532
Alexis R. Weber, I. Lednev
Luminescence spectroscopy is a versatile analytical technique that measures the emitted light resulted from the radiative deactivation of electronically excited states of molecular an atomic species. The field of forensic science has implemented the use of fluorescence spectroscopy for the analysis of bloodstains. Bloodstains discovered at crime scenes can provide crucial information to an investigation. It allows for the identification of the individual providing that there is a match with a known DNA profile. Additionally, determining the time since deposition (TSD) can assist investigators in establishing when the crime occurred or if a bloodstain present is related to the investigated event. However, most techniques that researchers have utilized thus far focus on the analysis of hemoglobin, both for identification and TSD determinations. Unlike other techniques, fluorescence spectroscopy can investigate the endogenous fluorophores within bloodstains. In this brief review, the ability of fluorescence spectroscopy for the analysis of bloodstains will be discussed. Including the ability to identify, determine the time since deposition, and phenotypic characterization of bloodstains.
{"title":"Brightness of blood: Review of fluorescence spectroscopy analysis of bloodstains","authors":"Alexis R. Weber, I. Lednev","doi":"10.3389/frans.2022.906532","DOIUrl":"https://doi.org/10.3389/frans.2022.906532","url":null,"abstract":"Luminescence spectroscopy is a versatile analytical technique that measures the emitted light resulted from the radiative deactivation of electronically excited states of molecular an atomic species. The field of forensic science has implemented the use of fluorescence spectroscopy for the analysis of bloodstains. Bloodstains discovered at crime scenes can provide crucial information to an investigation. It allows for the identification of the individual providing that there is a match with a known DNA profile. Additionally, determining the time since deposition (TSD) can assist investigators in establishing when the crime occurred or if a bloodstain present is related to the investigated event. However, most techniques that researchers have utilized thus far focus on the analysis of hemoglobin, both for identification and TSD determinations. Unlike other techniques, fluorescence spectroscopy can investigate the endogenous fluorophores within bloodstains. In this brief review, the ability of fluorescence spectroscopy for the analysis of bloodstains will be discussed. Including the ability to identify, determine the time since deposition, and phenotypic characterization of bloodstains.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-26","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43923143","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-07-13DOI: 10.3389/frans.2022.934639
Rushali Dargan, Clifford Samson, Wesley S. Burr, B. Daoust, S. Forbes
Cadaver detection dogs (CDDs) are trained to locate human remains and/or objects associated with human remains. This is possible due to their extraordinary olfactory capabilities compared to those of humans. To reinforce this capability, CDDs must be trained and regularly exposed to the target odor in the form of training aids which include—chemical formulations, animal remains, and/or human remains. Currently, the Ontario Provincial Police (OPP) use amputated limbs/feet from consented surgeries performed on diabetic patients as cadaver detection dog training aids. There is limited knowledge about the volatile organic compound (VOC) composition of these training aids and their appropriateness as an alternative to human remains for CDD training purposes, which formed the aim of the current study. VOCs were collected from amputated lower limbs/feet repeatedly using thermal desorption (TD) tubes and analyzed with comprehensive two-dimensional gas chromatography—time-of-flight mass spectrometry (GC×GC-TOFMS). The response of cadaver detection dogs to these training aids was also recorded to understand their alert in the context of the detected VOCs. VOC classes including acids, alcohols, aldehydes, ketones, ester and analogues, ethers, aliphatic, cyclics, sulfur-containing, nitrogen-containing, and halogen-containing VOCs were identified. Of these classes, cyclic VOCs were most abundant followed by nitrogen-containing VOCs while ethers were the least abundant. The most prominent VOCs identified in amputated limbs/feet were decomposition related however, one VOC—sevoflurane, originated from anaesthesia during the surgeries. It was determined that the VOC profile of aged and relatively recent training aids were variable. The aged training aids sampled over time had less variability (compared to more recent training aids). Additionally, the VOC profiles of samples was not found variable owing to the storage conditions—room temperature, refrigerator or freezer. Overall, a 98.4% detection rate was observed for amputated limbs/feet used as CDD training aids and the presence of non-decomposition related VOCs such as sevoflurane did not appear to impact the CDDs’ detection capability. This study highlights that the presence of decomposition VOCs in amputated limbs/feet and their high detection rate by CDDs validates their use as alternative CDD training aids.
{"title":"Validating the Use of Amputated Limbs Used as Cadaver Detection Dog Training Aids","authors":"Rushali Dargan, Clifford Samson, Wesley S. Burr, B. Daoust, S. Forbes","doi":"10.3389/frans.2022.934639","DOIUrl":"https://doi.org/10.3389/frans.2022.934639","url":null,"abstract":"Cadaver detection dogs (CDDs) are trained to locate human remains and/or objects associated with human remains. This is possible due to their extraordinary olfactory capabilities compared to those of humans. To reinforce this capability, CDDs must be trained and regularly exposed to the target odor in the form of training aids which include—chemical formulations, animal remains, and/or human remains. Currently, the Ontario Provincial Police (OPP) use amputated limbs/feet from consented surgeries performed on diabetic patients as cadaver detection dog training aids. There is limited knowledge about the volatile organic compound (VOC) composition of these training aids and their appropriateness as an alternative to human remains for CDD training purposes, which formed the aim of the current study. VOCs were collected from amputated lower limbs/feet repeatedly using thermal desorption (TD) tubes and analyzed with comprehensive two-dimensional gas chromatography—time-of-flight mass spectrometry (GC×GC-TOFMS). The response of cadaver detection dogs to these training aids was also recorded to understand their alert in the context of the detected VOCs. VOC classes including acids, alcohols, aldehydes, ketones, ester and analogues, ethers, aliphatic, cyclics, sulfur-containing, nitrogen-containing, and halogen-containing VOCs were identified. Of these classes, cyclic VOCs were most abundant followed by nitrogen-containing VOCs while ethers were the least abundant. The most prominent VOCs identified in amputated limbs/feet were decomposition related however, one VOC—sevoflurane, originated from anaesthesia during the surgeries. It was determined that the VOC profile of aged and relatively recent training aids were variable. The aged training aids sampled over time had less variability (compared to more recent training aids). Additionally, the VOC profiles of samples was not found variable owing to the storage conditions—room temperature, refrigerator or freezer. Overall, a 98.4% detection rate was observed for amputated limbs/feet used as CDD training aids and the presence of non-decomposition related VOCs such as sevoflurane did not appear to impact the CDDs’ detection capability. This study highlights that the presence of decomposition VOCs in amputated limbs/feet and their high detection rate by CDDs validates their use as alternative CDD training aids.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-07-13","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43457816","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-23DOI: 10.3389/frans.2022.930480
F. F. Sodré, Diogo de Jesus Soares Freire, Daniel B Alcântara, A. O. Maldaner
Cocaine and cannabis consumption during and after the 2019 Carnival holiday were assessed using the wastewater-based epidemiology (WBE) in the capital of Brazil, Brasília. The substances 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH), cocaine (COC), benzoylecgonine (BE), and cocaethylene (COE) were monitored in composite samples (24 h) collected in the entrance of North-Wing (NW) and South-Wing (SW) wastewater treatment plants (WWTP) for 15 consecutive days, including the Carnival holiday. Aliquots (100 ml) were enriched with isotope-labeled standards, solid-phase extracted and analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Results reveal higher cocaine consumption during the Carnival (average of 2.8 ± 0.7 g/1000inh/day) compared to the subsequent period (average of 1.7 ± 0.3 g/1000inh/day). Cannabis (THC) use was also higher during the holiday (14 ± 5 g/1000inh/day) but differences were not significative (unpaired t-test, 95%) compared to the following days (11 ± 3 g/1000inh/day), where consumption remained relatively constant corroborating that cannabis overall consumption is less affected by occasional abuse. Regarding cocaine, an unusual low consumption was noticed in the weekend immediately after the Carnival Holiday, indicating lower demand or supply issues. Higher cocaine and cannabis use was observed throughout the entire sampling period in the area covered by NW-WWTP, probably due to the higher proportion of young people. This investigation brings the first data on cannabis use in Brazil by WBE and confirms this strategy as a well consolidate tool for estimating illicit drug use and abuse.
{"title":"Understanding Illicit Drug Use Trends During the Carnival Holiday in the Brazilian Capital Through Wastewater Analysis","authors":"F. F. Sodré, Diogo de Jesus Soares Freire, Daniel B Alcântara, A. O. Maldaner","doi":"10.3389/frans.2022.930480","DOIUrl":"https://doi.org/10.3389/frans.2022.930480","url":null,"abstract":"Cocaine and cannabis consumption during and after the 2019 Carnival holiday were assessed using the wastewater-based epidemiology (WBE) in the capital of Brazil, Brasília. The substances 11-nor-9-carboxy-Δ9-tetrahydrocannabinol (THC-COOH), cocaine (COC), benzoylecgonine (BE), and cocaethylene (COE) were monitored in composite samples (24 h) collected in the entrance of North-Wing (NW) and South-Wing (SW) wastewater treatment plants (WWTP) for 15 consecutive days, including the Carnival holiday. Aliquots (100 ml) were enriched with isotope-labeled standards, solid-phase extracted and analyzed by liquid chromatography-tandem mass spectrometry (LC-MS/MS). Results reveal higher cocaine consumption during the Carnival (average of 2.8 ± 0.7 g/1000inh/day) compared to the subsequent period (average of 1.7 ± 0.3 g/1000inh/day). Cannabis (THC) use was also higher during the holiday (14 ± 5 g/1000inh/day) but differences were not significative (unpaired t-test, 95%) compared to the following days (11 ± 3 g/1000inh/day), where consumption remained relatively constant corroborating that cannabis overall consumption is less affected by occasional abuse. Regarding cocaine, an unusual low consumption was noticed in the weekend immediately after the Carnival Holiday, indicating lower demand or supply issues. Higher cocaine and cannabis use was observed throughout the entire sampling period in the area covered by NW-WWTP, probably due to the higher proportion of young people. This investigation brings the first data on cannabis use in Brazil by WBE and confirms this strategy as a well consolidate tool for estimating illicit drug use and abuse.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-06-23","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"44166282","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-09DOI: 10.3389/frans.2022.897605
F. Westad, Federico Marini
Variable selection is a topic of interest in many scientific communities. Within chemometrics, where the number of variables for multi-channel instruments like NIR spectroscopy and metabolomics in many situations is larger than the number of samples, the strategy has been to use latent variable regression methods to overcome the challenges with multiple linear regression. Thereby, there is no need to remove variables as such, as the low-rank models handle collinearity and redundancy. In most studies on variable selection, the main objective was to compare the prediction performance (RMSE or accuracy in classification) between various methods. Nevertheless, different methods with the same objective will, in most cases, give results that are not significantly different. In this study, we present three other main objectives: i) to eliminate variables that are not relevant; ii) to return a small subset of variables that has the same or better prediction performance as a model with all original variables; and iii) to investigate the consistency of these small subsets.
{"title":"Variable Selection and Redundancy in Multivariate Regression Models","authors":"F. Westad, Federico Marini","doi":"10.3389/frans.2022.897605","DOIUrl":"https://doi.org/10.3389/frans.2022.897605","url":null,"abstract":"Variable selection is a topic of interest in many scientific communities. Within chemometrics, where the number of variables for multi-channel instruments like NIR spectroscopy and metabolomics in many situations is larger than the number of samples, the strategy has been to use latent variable regression methods to overcome the challenges with multiple linear regression. Thereby, there is no need to remove variables as such, as the low-rank models handle collinearity and redundancy. In most studies on variable selection, the main objective was to compare the prediction performance (RMSE or accuracy in classification) between various methods. Nevertheless, different methods with the same objective will, in most cases, give results that are not significantly different. In this study, we present three other main objectives: i) to eliminate variables that are not relevant; ii) to return a small subset of variables that has the same or better prediction performance as a model with all original variables; and iii) to investigate the consistency of these small subsets.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-06-09","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49408204","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-06-08DOI: 10.3389/frans.2022.900483
Mitchell Tiessen, Holly M. Fruehwald, E. Easton, T. Stotesbury
Blood is an important type of forensic evidence because it can be used for source identification, toxicological analyses, and bloodstain pattern interpretation. Determining the time that bloodshed occurred, often described as the bloodstain’s time since deposition (TSD), has important implications for crime scene investigation. In this work, we focus on using electrochemical methods to monitor the gradual oxidative changes and electron-transfer reactions of hemoglobin (Hb) occurring in degrading bloodstains using differential pulse and hydrodynamic voltammetry. Bloodstains were monitored across a two-week time series in five different temperature conditions. Linear mixed models generated from the differential pulse voltammograms (DPV) suggested that 7 of 27 variables related to the redox reactions associated with the blood film were significantly correlated with time (p < 0.033). Of these correlated variables, all were related to the reduction of bound oxygen to hemoglobin or the oxidation of hemoglobin degradation products within the film. Hydrodynamic voltammetry demonstrated that hemoglobin retains its catalytic activity for oxygen reduction when aged on an electrode surface with a shift to greater peroxide formation the longer it is aged. The time series models are improved when the biological replicate is considered as a random effect, and as well as when peak area ratios are included in the model. Interestingly, using linear mixed models we observed a significant change in redox response at the 96-h time point (p < 0.043) regardless of temperature condition. Overall, we demonstrate preliminary support for DPV as a technique for TSD estimation of bloodstains.
{"title":"Insights Into Bloodstain Degradation and Time Since Deposition Estimation Using Electrochemistry","authors":"Mitchell Tiessen, Holly M. Fruehwald, E. Easton, T. Stotesbury","doi":"10.3389/frans.2022.900483","DOIUrl":"https://doi.org/10.3389/frans.2022.900483","url":null,"abstract":"Blood is an important type of forensic evidence because it can be used for source identification, toxicological analyses, and bloodstain pattern interpretation. Determining the time that bloodshed occurred, often described as the bloodstain’s time since deposition (TSD), has important implications for crime scene investigation. In this work, we focus on using electrochemical methods to monitor the gradual oxidative changes and electron-transfer reactions of hemoglobin (Hb) occurring in degrading bloodstains using differential pulse and hydrodynamic voltammetry. Bloodstains were monitored across a two-week time series in five different temperature conditions. Linear mixed models generated from the differential pulse voltammograms (DPV) suggested that 7 of 27 variables related to the redox reactions associated with the blood film were significantly correlated with time (p < 0.033). Of these correlated variables, all were related to the reduction of bound oxygen to hemoglobin or the oxidation of hemoglobin degradation products within the film. Hydrodynamic voltammetry demonstrated that hemoglobin retains its catalytic activity for oxygen reduction when aged on an electrode surface with a shift to greater peroxide formation the longer it is aged. The time series models are improved when the biological replicate is considered as a random effect, and as well as when peak area ratios are included in the model. Interestingly, using linear mixed models we observed a significant change in redox response at the 96-h time point (p < 0.043) regardless of temperature condition. Overall, we demonstrate preliminary support for DPV as a technique for TSD estimation of bloodstains.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-06-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49306755","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-25DOI: 10.3389/frans.2022.857880
R. J. Abel, J. Harynuk
Fire debris analysis is focused on the recovery and identification of ignitable liquids to provide context for fire investigation. Investigators use a variety of methods to select suspicious debris for analysis, with ignitable liquid detection canines being one of the most popular. When properly trained and certified, ignitable liquid detection canines offer continuous sampling with high sensitivity and the ability to discriminate between irrelevant and suspicious odours to rapidly locate debris which may contain ignitable liquid residues. However, canine indications are presumptive as they cannot be sufficiently scrutinised by the legal process without confirmatory laboratory analysis. Standard debris analysis methods detect very small amounts of ignitable liquid residue (∼1-0.1 μL) without maximising sensitivity which minimises the risk from false positives and from detection of background petroleum which is ubiquitous in our environment. For canine-selected debris, the goal of the laboratory analysis should be to provide data to confirm or refute the validity of the canine indication. For such confirmatory analysis to be useful, analytical sensitivity should approximate the sensitivity of the canine. The sensitivity of fire debris analysis is most influenced by the selection of the extraction device and tuning of extraction conditions. Non-destructive extractions are preferred for forensic analyses, and solid phase microextraction (SPME) offers an excellent option. However, the original SPME fibres are fragile and tend to skew the chromatographic profile which can lead to high costs and a risk of ignitable liquid misclassification. Herein, we present an optimised SPME extraction method suited to confirmatory analysis of canine-selected exhibits. The method is non-destructive and non-exhaustive, is easily applied to cans of debris, and yields chromatographic profiles equivalent to those obtained by the gold-standard passive headspace sampling (PHS) methods based on activated carbon. Fibre selection, debris temperature, fibre temperature, and extraction time were optimised to yield chromatographic profiles with maximum comparability to reference samples collected as neat liquids or standard PHS extracts. The optimised method is applied to samples recovered from another study which estimated the threshold of the canine’s sensitivity, with the laboratory result compared to the canine result for each sample.
{"title":"Sensitive and Representative Extraction of Petroleum-Based Ignitable Liquids From Fire Debris For Confirmatory Analysis of Canine-Selected Exhibits","authors":"R. J. Abel, J. Harynuk","doi":"10.3389/frans.2022.857880","DOIUrl":"https://doi.org/10.3389/frans.2022.857880","url":null,"abstract":"Fire debris analysis is focused on the recovery and identification of ignitable liquids to provide context for fire investigation. Investigators use a variety of methods to select suspicious debris for analysis, with ignitable liquid detection canines being one of the most popular. When properly trained and certified, ignitable liquid detection canines offer continuous sampling with high sensitivity and the ability to discriminate between irrelevant and suspicious odours to rapidly locate debris which may contain ignitable liquid residues. However, canine indications are presumptive as they cannot be sufficiently scrutinised by the legal process without confirmatory laboratory analysis. Standard debris analysis methods detect very small amounts of ignitable liquid residue (∼1-0.1 μL) without maximising sensitivity which minimises the risk from false positives and from detection of background petroleum which is ubiquitous in our environment. For canine-selected debris, the goal of the laboratory analysis should be to provide data to confirm or refute the validity of the canine indication. For such confirmatory analysis to be useful, analytical sensitivity should approximate the sensitivity of the canine. The sensitivity of fire debris analysis is most influenced by the selection of the extraction device and tuning of extraction conditions. Non-destructive extractions are preferred for forensic analyses, and solid phase microextraction (SPME) offers an excellent option. However, the original SPME fibres are fragile and tend to skew the chromatographic profile which can lead to high costs and a risk of ignitable liquid misclassification. Herein, we present an optimised SPME extraction method suited to confirmatory analysis of canine-selected exhibits. The method is non-destructive and non-exhaustive, is easily applied to cans of debris, and yields chromatographic profiles equivalent to those obtained by the gold-standard passive headspace sampling (PHS) methods based on activated carbon. Fibre selection, debris temperature, fibre temperature, and extraction time were optimised to yield chromatographic profiles with maximum comparability to reference samples collected as neat liquids or standard PHS extracts. The optimised method is applied to samples recovered from another study which estimated the threshold of the canine’s sensitivity, with the laboratory result compared to the canine result for each sample.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-05-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45995684","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-05-19DOI: 10.3389/frans.2022.867938
Michael D. Sorochan Armstrong, A. P. de la Mata, J. Harynuk
Discriminant-type analyses arise from the need to classify samples based on their measured characteristics (variables), usually with respect to some observable property. In the case of samples that are difficult to obtain, or using advanced instrumentation, it is very common to encounter situations with many more measured characteristics than samples. The method of Partial Least Squares Regression (PLS-R), and its variant for discriminant-type analyses (PLS-DA) are among the most ubiquitous of these tools. PLS utilises a rank-deficient method to solve the inverse least-squares problem in a way that maximises the co-variance between the known properties of the samples (commonly referred to as the Y-Block), and their measured characteristics (the X-block). A relatively small subset of highly co-variate variables are weighted more strongly than those that are poorly co-variate, in such a way that an ill-posed matrix inverse problem is circumvented. Feature selection is another common way of reducing the dimensionality of the data to a relatively small, robust subset of variables for use in subsequent modelling. The utility of these features can be inferred and tested any number of ways, this are the subject of this review.
{"title":"Review of Variable Selection Methods for Discriminant-Type Problems in Chemometrics","authors":"Michael D. Sorochan Armstrong, A. P. de la Mata, J. Harynuk","doi":"10.3389/frans.2022.867938","DOIUrl":"https://doi.org/10.3389/frans.2022.867938","url":null,"abstract":"Discriminant-type analyses arise from the need to classify samples based on their measured characteristics (variables), usually with respect to some observable property. In the case of samples that are difficult to obtain, or using advanced instrumentation, it is very common to encounter situations with many more measured characteristics than samples. The method of Partial Least Squares Regression (PLS-R), and its variant for discriminant-type analyses (PLS-DA) are among the most ubiquitous of these tools. PLS utilises a rank-deficient method to solve the inverse least-squares problem in a way that maximises the co-variance between the known properties of the samples (commonly referred to as the Y-Block), and their measured characteristics (the X-block). A relatively small subset of highly co-variate variables are weighted more strongly than those that are poorly co-variate, in such a way that an ill-posed matrix inverse problem is circumvented. Feature selection is another common way of reducing the dimensionality of the data to a relatively small, robust subset of variables for use in subsequent modelling. The utility of these features can be inferred and tested any number of ways, this are the subject of this review.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-05-19","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48894428","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Pub Date : 2022-04-29DOI: 10.3389/frans.2022.868515
T. Huck, Raphaël Bajon, N. Grima, E. Portela, J. Molines, T. Penduff
The fate of plastics entering the 3D ocean circulation from rivers discharge is examined through the Lagrangian analysis of neutrally buoyant particles. Particles are released continuously over 1991–2010 at the surface along the coasts according to monthly estimates of rivers plastic waste input. They are advected by daily currents from a state-of-the-art global ocean model at 1/12° resolution. At the end of the simulation (year 2010), particles remaining in the surface layer of 1 m thickness represent less than 2% of the total particles released. These are concentrated in the center of subtropical gyres, mostly in the South Indian Ocean, and the North Pacific, in relation with the large sources from Asia, and in good agreement with previous 2D numerical experiments in the surface layer. These patterns remain similar down to about 30 m depth, this upper layer strongly influenced by Ekman currents trapping about 20% of the total released particles. About 50% of the total released particles remain in the upper 100 m, and up to 90% are found in the upper 400 m at the end of the experiment. Below the mixed layer, they are more widely dispersed horizontally and follow the main global pathways of ocean ventilation of mode and deep water masses. Plastic particles, neutrally buoyant because of their small size or biofouling, are thus expected to be strongly dispersed in the global ocean thermocline following mode waters patterns, and reach the deeper layers following the North Atlantic Deep Water formation path. Two major source regions have a global impact. Particles from the western North Pacific spread over the whole Pacific Ocean poleward of 20°S, whereas particles from Indonesia spread over the whole latitude band from 60°S to 20°S.
{"title":"Three-Dimensional Dispersion of Neutral “Plastic” Particles in a Global Ocean Model","authors":"T. Huck, Raphaël Bajon, N. Grima, E. Portela, J. Molines, T. Penduff","doi":"10.3389/frans.2022.868515","DOIUrl":"https://doi.org/10.3389/frans.2022.868515","url":null,"abstract":"The fate of plastics entering the 3D ocean circulation from rivers discharge is examined through the Lagrangian analysis of neutrally buoyant particles. Particles are released continuously over 1991–2010 at the surface along the coasts according to monthly estimates of rivers plastic waste input. They are advected by daily currents from a state-of-the-art global ocean model at 1/12° resolution. At the end of the simulation (year 2010), particles remaining in the surface layer of 1 m thickness represent less than 2% of the total particles released. These are concentrated in the center of subtropical gyres, mostly in the South Indian Ocean, and the North Pacific, in relation with the large sources from Asia, and in good agreement with previous 2D numerical experiments in the surface layer. These patterns remain similar down to about 30 m depth, this upper layer strongly influenced by Ekman currents trapping about 20% of the total released particles. About 50% of the total released particles remain in the upper 100 m, and up to 90% are found in the upper 400 m at the end of the experiment. Below the mixed layer, they are more widely dispersed horizontally and follow the main global pathways of ocean ventilation of mode and deep water masses. Plastic particles, neutrally buoyant because of their small size or biofouling, are thus expected to be strongly dispersed in the global ocean thermocline following mode waters patterns, and reach the deeper layers following the North Atlantic Deep Water formation path. Two major source regions have a global impact. Particles from the western North Pacific spread over the whole Pacific Ocean poleward of 20°S, whereas particles from Indonesia spread over the whole latitude band from 60°S to 20°S.","PeriodicalId":73063,"journal":{"name":"Frontiers in analytical science","volume":null,"pages":null},"PeriodicalIF":0.0,"publicationDate":"2022-04-29","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"49216406","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
京公网安备 11010802042870号
京ICP备2023020795号-1
扫码关注我们
求助内容：
应助结果提醒方式：