Enhancing skin lesion classification with advanced deep learning ensemble models: a path towards accurate medical diagnostics.

Skin cancer, including the highly lethal malignant melanoma, poses a significant global health challenge with a rising incidence rate. Early detection plays a pivotal role in improving survival rates. This study aims to develop an advanced deep learning-based approach for accurate skin lesion classification, addressing challenges such as limited data availability, class imbalance, and noise. Modern deep neural network designs, such as ResNeXt101, SeResNeXt101, ResNet152V2, DenseNet201, GoogLeNet, and Xception, which are used in the study and ze optimised using the SGD technique. The dataset comprises diverse skin lesion images from the HAM10000 and ISIC datasets. Noise and artifacts are tackled using image inpainting, and data augmentation techniques enhance training sample diversity. The ensemble technique is utilized, creating both average and weighted average ensemble models. Grid search optimizes model weight distribution. The individual models exhibit varying performance, with metrics including recall, precision, F1 score, and MCC. The "Average ensemble model" achieves harmonious balance, emphasizing precision, F1 score, and recall, yielding high performance. The "Weighted ensemble model" capitalizes on individual models' strengths, showcasing heightened precision and MCC, yielding outstanding performance. The ensemble models consistently outperform individual models, with the average ensemble model attaining a macro-average ROC-AUC score of 96 % and the weighted ensemble model achieving a macro-average ROC-AUC score of 97 %. This research demonstrates the efficacy of ensemble techniques in significantly improving skin lesion classification accuracy. By harnessing the strengths of individual models and addressing their limitations, the ensemble models exhibit robust and reliable performance across various metrics. The findings underscore the potential of ensemble techniques in enhancing medical diagnostics and contributing to improved patient outcomes in skin lesion diagnosis.

Novel Insights into the Bioactive Metabolites of Macrocybe gigantea (Agaricomycetes), Using Gas Chromatography-Mass Spectrometry Combined with Chemoinformatic Approaches.

Macrocybe gigantea is an edible mushroom and has multiple pharmacological properties, including antibacterial, antioxidant, and antitumor activities. However, only a few reports are currently available on the bioactive compounds and bioactivity of this mushroom. Therefore, the present study aimed to explore the unique chemical diversity of the fruiting body of M. gigantea. Species identification was done accurately with morphological and molecular methods, followed by mycochemical extraction in different solvent systems. The ethanolic extract of the fruiting body gave maximum yield, and gas chromatography-mass spectrometry (GC-MS) analysis was performed along with an assessment of antibacterial activity and cell viability by the MTT assay. The GC-MS analysis revealed 50 metabolites, and further cheminformatics analysis of these metabolites revealed their possible biological activities. In addition, the physicochemical and mineral element analysis of M. gigantea revealed the quality and authenticity of the species. Altogether, the current investigation gives a comprehensive overview of the bioactive metabolites of M. gigantea.

Inhibition of DNA amplification caused by metal in extracted bloodstains and in direct amplification.

Copper is a metal that is commonly found in many evidence types, such as weapons, and household items, that are often retrieved from crime scenes. We used copper and brass sheets, cartridge cases, and metal décor items as substrates for deposition of blood to study inhibition induced by metals. Another objective was to compare the efficiency of STR profiling following extraction, quantification, amplification, and capillary electrophoresis with a direct amplification method which only requires amplification and capillary electrophoresis. Samples that were designated for extraction, were collected, extracted, and purified using the Qiagen EZ1 DNA Investigator® Kit. Direct amplification was performed using Copan microFLOQ® Direct Swabs to collect minute amount of blood. Complete STR profiles were obtained from 54% of the extracted samples and from 88% of directly amplified samples. Substrate type and amplification method (two-way ANOVA, F(5, 131) = 3.59, p = 4.50 × 10-3) impacted profile completeness but collection time did not (one-way ANOVA, F(1, 131) = 0.017, p = 0.90). Composition of the substrates was determined with scanning electron microscopy (SEM), and inhibition caused by copper was evaluated for each substrate. When substrate composition was analyzed in combination with amplification method, a substrate's copper percent was not associated with increased allelic dropout (one-way ANOVA, F(1, 176) = 0.70, p = 0.40) but amplification alone was statistically significant (one-way ANOVA, F(1, 176) = 40.64, p = 1.56 × 10-9). Using the Copan microFLOQ® Direct Swabs, inhibition induced by metal observed was not as pronounced. In addition, less variability in profile completeness was detected, profiles were generated within a very short period, and only minute amount of bloodstain was needed, allowing for additional analysis of the same samples. Our results directly quantify the effect of copper composition on DNA profiling and suggest that while copper percentage is an observational contributor to allele dropout it is not wholly responsible for these events.

High-throughput DNA sequencing of environmentally insulted latent fingerprints after visualization with nanoscale columnar-thin-film technique.

The goals of this study were to (a) ascertain human identity capabilities of DNA obtained from latent fingerprints that have been first environmentally insulted and then developed by the deposition of a columnar thin film (CTF), and (b) to determine if the CTF process and material are detrimental to single nucleotide polymorphism (SNP) analysis. Fingerprints were deposited on five different types of substrates and aged for one day, 7 days or 30 days while being environmentally insulted under one of the four conditions: 16.6 °C and 60% relative humidity (RH) (Condition A), 24.5 °C and 60% RH (Condition B), 35 °C and 67% RH (Condition C) and a cold condition (Condition D). Then CTF technique was then on 59% of these fingerprints. DNA samples from 805 fingerprints were extracted, quantified, subjected to manual library preparation using the Precision ID Identity Panel, and underwent high-throughput sequencing. The Ion S5™ platform was employed to sequence 124 SNP amplicons. SNPs were successfully sequenced from 802/805 samples. Total read depth was consistent across environmental conditions, and majority of samples had 100% profile completeness and 100% concordance. Anecdotally, libraries that were amplified with a higher cycle number had more 'Major Allele Frequency' flags compared to samples amplified with 23 cycle numbers, possibly due to stochastic effects. Neither the substrates nor the CTF process and materials inhibit downstream DNA analysis. DNA of low quality and quantity from the chosen samples can be sequenced using the Precision ID Identity Panel on the Ion S5™ platform which performed well, however, a different approach may be needed if spurious alleles are suspected.

Massively parallel sequencing and STR analysis from partial bloody fingerprints enhanced with columnar thin films.

Fingerprint enhancement often includes either physical or chemical approaches, such as fingerprint powder or cyanoacrylate fuming, to improve the quality of a fingerprint for visualization and analysis. However, these methods become more complex when fingerprints are partial bloody, and these procedures may interfere with downstream DNA analysis. Columnar thin film (CTF) deposition is a type of nanotechnology that utilizes an evaporant material to enhance a fingerprint under low-pressure conditions. Short tandem repeat (STR) analysis is the traditional method employed in crime laboratories. When DNA is of poor quality and quantity, like that often obtained from fingerprints, little to no genetic information may be obtained. Single nucleotide polymorphisms (SNPs) may be used to glean additional information when STR analysis fails. In this pilot study, 100 partial bloody fingerprints were collected from two donors and deposited on five different crime scene substrates, in which half were enhanced with CTFs and were graded for quality by an IAI-certified latent fingerprint examiner. CTF-developed fingerprints, on average, had higher grades compared to non-developed partial bloody fingerprints. STR analysis using Fusion 6C was performed to assess inhibition from the evaporant materials, in which no inhibition was observed. Sequencing of SNPs using the Precision ID Identity Panel was also employed, in which genetic information that could not be obtained from STRs was acquired with SNPs. Various sample types (i.e. pristine, low quality, and contaminated) utilized in this project demonstrated the acceptable performance of the Precision ID Identity Panel.

Determination of human identity from Anopheles stephensi mosquito blood meals using direct amplification and massively parallel sequencing.

DNA obtained from biological evidence can link individuals to a crime scene. DNA is typically obtained from body fluids deposited on various substrates such as fabric or common household objects. However, other unusual sources of human biological material can also be used to generate DNA profiles. Here we show that short tandem repeat (STR) DNA profiles can also be obtained from single source and mixtures of human DNA in the blood meals of Anopheles stephensi mosquitoes. Using direct amplification with the PowerPlex® Fusion 6C System, we have determined that full and partial profiles can be obtained by assessing degradation of DNA at various times post-feed up to 20-24 h post-blood meal. Moreover, we can assign donor identity through both STR profiles, as well as through single nucleotide polymorphisms (SNPs) detected using massively parallel sequencing (MPS) with the Precision ID Identity Panel and Ion Chef™/Ion S5™ System up to 24-48 h post-blood meal. Based on the results from a total of 490 mosquitoes fed on 11 different sources of human blood, we conclude that both STR and SNP technologies can be applied to mosquito blood meals as effective forensic approaches to determine the identity of specific individuals and establish the timing of their presence at a crime scene.

Comparison of Quantifiler(®) Trio and InnoQuant™ human DNA quantification kits for detection of DNA degradation in developed and aged fingerprints.

The development techniques employed to visualize fingerprints collected from crime scenes as well as post-development ageing may result in the degradation of the DNA present in low quantities in such evidence samples. Amplification of the DNA samples with short tandem repeat (STR) amplification kits may result in partial DNA profiles. A comparative study of two commercially available quantification kits, Quantifiler(®) Trio and InnoQuant™, was performed on latent fingerprint samples that were either (i) developed using one of three different techniques and then aged in ambient conditions or (ii) undeveloped and then aged in ambient conditions. The three fingerprint development techniques used were: cyanoacrylate fuming, dusting with black powder, and the columnar-thin-film (CTF) technique. In order to determine the differences between the expected quantities and actual quantities of DNA, manually degraded samples generated by controlled exposure of DNA standards to ultraviolet radiation were also analyzed. A total of 144 fingerprint and 42 manually degraded DNA samples were processed in this study. The results indicate that the InnoQuant™ kit is capable of producing higher degradation ratios compared to the Quantifiler(®) Trio kit. This was an expected result since the degradation ratio is a relative value specific for a kit based on the length and extent of amplification of the two amplicons that vary from one kit to the other. Additionally, samples with lower concentrations of DNA yielded non-linear relationships of degradation ratio with the duration of aging, whereas samples with higher concentrations of DNA yielded quasi-linear relationships. None of the three development techniques produced a noticeably different degradation pattern when compared to undeveloped fingerprints, and therefore do not impede downstream DNA analysis.

Quantitative and qualitative study of STR DNA from ethanol and formalin fixed tissues.

Complete and concordant autosomal short tandem repeat (STR) DNA profiles were obtained from 2.0mg human tissue samples of various types after they were preserved for 24 weeks in 100% ethanol and amplified with the GlobalFiler(®) and the PowerPlex(®) Fusion Amplification Kits. When 4.0mg of the same tissues were preserved for 12 weeks in 10% Neutral Buffered Formalin (NBF) they yielded partial profiles when amplified with the same kits. However, these NBF preserved tissues yielded complete autosomal profiles when amplified with the AmpFlSTR(®) MiniFiler™ Amplification Kit. Six tissue specimens from the male donor were also amplified with the PowerPlex(®) Y-23 System. Y-STR profiles were successfully generated from 2.0mg tissue specimens when preserved for 12 weeks in 100% Ethanol. Only partial profiles were obtained when the fixation time was increased to 24 weeks. Only partial Y-STR profiles were also obtained from 4.0mg tissue specimen from the same donor when preserved in 10% NBF. In an attempt to optimize the method, the preserved samples that yielded partial profiles were homogenized using the BioMasher III disposable homogenizer and BioMasher III homogenizer and filter. These homogenized tissues did not yield significantly better or more complete profiles when using the GlobalFiler(®), AmpFlSTR(®) MiniFiler™ Amplification Kits, the PowerPlex(®) Fusion System or the PowerPlex(®) Y23 System. A total number of 240 tissue samples were analyzed in this project. The amplification of the tissues preserved in 10% NBF with kits such as AmpFlSTR(®) MiniFiler™ and GlobalFiler(®) Amplification Kits that contain mini STR primers can be beneficial in forensic testing. The results of the study indicate that in cases such as when a victim or a suspect is missing, the profiles obtained from minute amounts of chemically fixed tissues can be used as reference samples and compared to evidence found at the crime scene.

Direct Y-STR amplification of body fluids deposited on commonly found crime scene substrates.

Body fluids detected on commonly found crime scene substrates require extraction, purification and quantitation of DNA prior to amplification and generation of short tandem repeat (STR) DNA profiles. In this research Y-STR profiles were generated via direct amplification of blood and saliva deposited on 12 different substrates. These included cigarette butts, straws, grass, leaves, woodchips and seven different types of fabric. After depositing either 0.1 µL of blood or 0.5 µL of saliva, each substrate containing the dry body fluid stain was punched using a Harris 1.2 mm micro-punch. Each of these punched substrates, a total of 720 samples, containing minute amount of blood or saliva was either amplified directly without any pre-treatment, or was treated with one of the four washing reagents or buffer. In each of these five experimental groups the substrates containing the body fluid remained in the amplification reagent during the thermal cycling process. Each sample was amplified with the three direct Y-STR amplification kits; AmpFℓSTR(®) Yfiler(®) Direct, Yfiler(®) Plus Amplification Kits and the PowerPlex(®) Y23 System. Complete and concordant Y-STR profiles were successfully obtained from most of these 12 challenging crime scene objects when the stains were analyzed by at least one of the five experimental groups. The reagents and buffer were interchangeable among the three amplification kits, however, pre-treatment with these solutions did not appear to enhance the quality or the number of the full profiles generated with direct amplification. This study demonstrates that blood and saliva deposited on these simulated crime scene objects can be amplified directly.

Generation of DNA profiles from fingerprints developed with columnar thin film technique.

Partial-bloody fingerprints and partial fingerprints with saliva are often encountered at crime scenes, potentially enabling the combination of fingerprint and DNA analyses for absolute identification, provided that the development technique for fingerprint analysis does not inhibit DNA analysis. 36 partial-bloody fingerprints and 30 fingerprints wetted with saliva, all deposited on brass, were first developed using the columnar-thin-film (CTF) technique and then subjected to short tandem repeat (STR) DNA analysis. Equal numbers of samples were subjected to the same DNA analysis without development. Tris (8-hydroxyquinolinato) aluminum, or Alq3, was evaporated to deposit CTFs for development of the prints. DNA was extracted from all 132 samples, quantified, and amplified with AmpFlSTR(®) Identifiler Plus Amplification Kit. Additionally, DNA analyses were conducted on four blood smears on un-fingerprinted brass that had been subjected to CTF deposition and four blood smears on un-fingerprinted brass that had not been subjected to CTF deposition. Complete and concordant autosomal STR profiles of the same quality were obtained from both undeveloped and CTF-developed fingerprints, indicating that CTF development of fingerprints preserves DNA and does not inhibit subsequent DNA analysis. Even when there were no fingerprints, CTF deposition did not lead to inhibition of DNA analysis.

Evaluation of Direct PCR Amplification Using Various Swabs and Washing Reagents.

DNA profiles were generated via direct amplification from blood and saliva samples deposited on various types of swab substrates. Each of the six non-FTA substrates used in this research was punched with a Harris 1.2 mm puncher. After 0.1 µL of blood or 0.5 µL saliva, samples were deposited on each of these punches, samples were pretreated with one of four buffers and washing reagents. Amplification was performed using direct and nondirect autosomal and Y-STR kits. Autosomal and Y-STR profiles were successfully generated from most of these substrates when pretreated with buffer or washing reagents. Concordant profiles were obtained within and between the six substrates, the six amplification kits, and all four reagents. The direct amplification of substrates which do not contain lysing agent would be beneficial to the forensic community as the procedure can be used on evidence samples commonly found at crime scenes.

Detection of insertion/deletion polymorphisms from challenged samples using the Investigator DIPplex® kit.

This research focuses on detection of bi-allelic insertion/deletion polymorphisms (InDels) from challenged samples using the Investigator DIPplex® Kit from Qiagen. The study included analyzing body fluids from humans, as well as pristine and degraded samples. For the purpose of assessing species specificity, samples from various animals were included. At first, an analytical threshold (AT) for the detection of alleles was established based on an assessment of the noise in the system. Then, InDel profiles were obtained from samples exposed to detrimental environmental conditions, washed bloodstains, lipsticks, ChapStick®, ancient Croatian bone samples, and every day products such as toothbrushes and dental floss. Concordant profiles were obtained from different body fluids of the same donor. InDel profiles were also generated successfully when body fluids were deposited on substrates and directly amplified without pre-treatment with buffer or washing reagents. InDels can provide additional information when only partial STR profiles are generated from challenged samples.

An evaluation of direct PCR amplification.

AIM: To generate complete DNA profiles from blood and saliva samples deposited on FTA® and non-FTA® paper substrates following a direct amplification protocol. METHODS: Saliva samples from living donors and blood samples from deceased individuals were deposited on ten different FTA® and non-FTA® substrates. These ten paper substrates containing body fluids were kept at room temperature for varying lengths of time ranging from one day to approximately one year. For all assays in this research, 1.2 mm punches were collected from each substrate containing one type of body fluid and amplified with reagents provided in the nine commercial polymerase chain reaction (PCR) amplification kits. The substrates were not subjected to purification reagent or extraction buffer prior to amplification. RESULTS: Success rates were calculated for all nine amplification kits and all ten substrates based on their ability to yield complete DNA profiles following a direct amplification protocol. Six out of the nine amplification kits, and four out of the ten paper substrates had the highest success rates overall. CONCLUSION: The data show that it is possible to generate complete DNA profiles following a direct amplification protocol using both standard (non-direct) and direct PCR amplification kits. The generation of complete DNA profiles appears to depend more on the success of the amplification kit rather than the than the FTA®- or non-FTA®-based substrates.

Generating DNA profiles from immunochromatographic cards using LCN methodology.

The aim of this research was to obtain DNA profiles from immunochromatographic test devices which have already yielded positive results with body fluids obtained from fourteen volunteers. Three different immunochromatographic cards for the identification of human blood and one for the identification of human saliva were used for this research. Each body fluid was detected using the appropriate immunochromatographic card. The used cards were kept at room temperature for various lengths of time. The membranes were removed at the end of the designated times and the entire strip was extracted using low copy number (LCN) extraction procedure. The extracted DNA was amplified using reduced amplification volume and higher PCR cycle numbers. Autosomal STR profiles were detected using AmpFℓSTR(®) Identifiler™ PCR Amplification Kit from Applied Biosystems (AB). Additionally, DNA extracted from the male volunteers was amplified using the AB AmpFℓSTR(®) Yfiler™ PCR Amplification Kit. Analysis of the amplified products was carried out by capillary electrophoresis injection on the AB 3130xl Genetic Analyzer. The generated DNA data was analyzed using the SoftGenetics GeneMarker(®) HID Version 1.7 software. Autosomal and Y-STR DNA profiles were obtained from most of the cards which were stored at room temperature for up to three months. DNA profile was obtained from all four types of the immunochromatographic cards used in this study. These profiles were concordant with the profiles obtained from the donors' reference samples.

Analysis of human fecal material for autosomal and Y chromosome STRs.

Human stool samples from eight volunteers were stored under various conditions and extracted by three different procedures. Fecal material and tissue paper soiled with fecal material obtained from a crime scene were also extracted. Extracted DNA was amplified using the AmpFlSTR Profiler Plus, AmpFlSTR COfiler, and the AmpFlSTR Identifiler PCR amplification kits for the detection of the autosomal STR allelic patterns. DNA extracted from the male volunteers and from the soiled tissue paper evidence sample was also amplified using the Y-PLEX 6 and Y-PLEX 5 amplification kits. Analysis of the amplified products was carried out by capillary electrophoresis on the ABI PRISM 310 Genetic Analyzer. Autosomal and Y-STR profiles obtained from the fecal material were concordant with the results from the donors' buccal swabs.

Human immunodeficiency virus type 1 infection in twin pairs infected at birth.

Host genetic factors may influence the course of human immunodeficiency virus (HIV) infection. In Blantyre, Malawi, polymerase chain reaction was used to identify twin pairs who were concordantly HIV-1-infected in utero or perinatally and then to examine strain divergence or virus levels in identical and fraternal twin pairs. Among 315 twin pairs, both infants in 14 fraternal and 5 identical pairs were found to be infected at the same visit. Among 10 pairs, HIV-1 sequences were determined for both infants at >or=1 time point. HIV levels had a common profile in both fraternal and identical twin pairs. Identical twins were not always infected by the same quasi species, indicating that their mothers had multiple quasi species capable of infecting their infants. Subsequent viral divergence appears to depend on quasi-species stability rather than on genetically controlled host immune responses. Thus, given infection, factors intrinsic to HIV-1 are more important than host genetics in viral evolution.