Weakly-supervised deep learning models enable HER2-low prediction from H &E stained slides.

BACKGROUND: Human epidermal growth factor receptor 2 (HER2)-low breast cancer has emerged as a new subtype of tumor, for which novel antibody-drug conjugates have shown beneficial effects. Assessment of HER2 requires several immunohistochemistry tests with an additional in situ hybridization test if a case is classified as HER2 2+. Therefore, novel cost-effective methods to speed up the HER2 assessment are highly desirable. METHODS: We used a self-supervised attention-based weakly supervised method to predict HER2-low directly from 1437 histopathological images from 1351 breast cancer patients. We built six distinct models to explore the ability of classifiers to distinguish between the HER2-negative, HER2-low, and HER2-high classes in different scenarios. The attention-based model was used to comprehend the decision-making process aimed at relevant tissue regions. RESULTS: Our results indicate that the effectiveness of classification models hinges on the consistency and dependability of assay-based tests for HER2, as the outcomes from these tests are utilized as the baseline truth for training our models. Through the use of explainable AI, we reveal histologic patterns associated with the HER2 subtypes. CONCLUSION: Our findings offer a demonstration of how deep learning technologies can be applied to identify HER2 subgroup statuses, potentially enriching the toolkit available for clinical decision-making in oncology.

First Transcriptome Analysis of Hepatoblastoma in Brazil: Unraveling the Pivotal Role of Noncoding RNAs and Metabolic Pathways.

Hepatoblastoma stands as the most prevalent liver cancer in the pediatric population. Characterized by a low mutational burden, chromosomal and epigenetic alterations are key drivers of its tumorigenesis. Transcriptome analysis is a powerful tool for unraveling the molecular intricacies of hepatoblastoma, shedding light on the effects of genetic and epigenetic changes on gene expression. In this study conducted in Brazilian patients, an in-depth whole transcriptome analysis was performed on 14 primary hepatoblastomas, compared to control liver tissues. The analysis unveiled 1,492 differentially expressed genes (1,031 upregulated and 461 downregulated), including 920 protein-coding genes (62%). Upregulated biological processes were linked to cell differentiation, signaling, morphogenesis, and development, involving known hepatoblastoma-associated genes (DLK1, MEG3, HDAC2, TET1, HMGA2, DKK1, DKK4), alongside with novel findings (GYNG4, CDH3, and TNFRSF19). Downregulated processes predominantly centered around oxidation and metabolism, affecting amines, nicotinamides, and lipids, featuring novel discoveries like the repression of SYT7, TTC36, THRSP, CCND1, GCK and CAMK2B. Two genes, which displayed a concordant pattern of DNA methylation alteration in their promoter regions and dysregulation in the transcriptome, were further validated by RT-qPCR: the upregulated TNFRSF19, a key gene in the embryonic development, and the repressed THRSP, connected to lipid metabolism. Furthermore, based on protein-protein interaction analysis, we identified genes holding central positions in the network, such as HDAC2, CCND1, GCK, and CAMK2B, among others, that emerged as prime candidates warranting functional validation in future studies. Notably, a significant dysregulation of non-coding RNAs (ncRNAs), predominantly upregulated transcripts, was observed, with 42% of the top 50 highly expressed genes being ncRNAs. An integrative miRNA-mRNA analysis revealed crucial biological processes associated with metabolism, oxidation reactions of lipids and carbohydrates, and methylation-dependent chromatin silencing. In particular, four upregulated miRNAs (miR-186, miR-214, miR-377, and miR-494) played a pivotal role in the network, potentially targeting multiple protein-coding transcripts, including CCND1 and CAMK2B. In summary, our transcriptome analysis highlighted disrupted embryonic development as well as metabolic pathways, particularly those involving lipids, emphasizing the emerging role of ncRNAs as epigenetic regulators in hepatoblastomas. These findings provide insights into the complexity of the hepatoblastoma transcriptome and identify potential targets for future therapeutic interventions.

Germline variants in early and late-onset Brazilian prostate cancer patients.

BACKGROUND: The median age for Prostate Cancer (PCa) diagnosis is 66 years, but 10% are diagnosed before 55 years. Studies on early-onset PCa remain both limited and controversial. This investigation sought to identify and characterize germline variants within Brazilian PCa patients classified as either early or later onset disease. METHODS: Peripheral blood DNA from 71 PCa patients: 18 younger (≤ 55 years) and 53 older (≥ 60 years) was used for Targeted DNA sequencing of 20 genes linked to DNA damage response, transcriptional regulation, cell cycle, and epigenetic control. Subsequent genetic variant identification was performed and variant functional impacts were analyzed with in silico prediction. RESULTS: A higher frequency of variants in the BRCA2 and KMT2C genes across both age groups. KMT2C has been linked to the epigenetic dysregulation observed during disease progression in PCa. We present the first instance of KMT2C mutation within the blood of Brazilian PCa patients. Furthermore, out of the recognized variants within the KMT2C gene, 7 were designated as deleterious. Thirteen deleterious variants were exclusively detected in the younger group, while the older group exhibited 37 variants. Within these findings, 4 novel variants emerged, including 1 designated as pathogenic. CONCLUSIONS: Our findings contribute to a deeper understanding of the genetic factors associated with PCa susceptibility in different age groups, especially among the Brazilian population. This is the first investigation to explore germline variants specifically in younger Brazilian PCa patients, with high relevance given the genetic diversity of the population in Brazil. Additionally, our work presents evidence of functionally deleterious germline variants within the KMT2C gene among Brazilian PCa patients. The identification of novel and functionally significant variants in the KMT2C gene emphasizes its potential role in PCa development and warrants further investigation.

Tracking SARS-CoV-2 variants of concern in wastewater: an assessment of nine computational tools using simulated genomic data.

Wastewater-based surveillance (WBS) is an important epidemiological and public health tool for tracking pathogens across the scale of a building, neighbourhood, city, or region. WBS gained widespread adoption globally during the SARS-CoV-2 pandemic for estimating community infection levels by qPCR. Sequencing pathogen genes or genomes from wastewater adds information about pathogen genetic diversity, which can be used to identify viral lineages (including variants of concern) that are circulating in a local population. Capturing the genetic diversity by WBS sequencing is not trivial, as wastewater samples often contain a diverse mixture of viral lineages with real mutations and sequencing errors, which must be deconvoluted computationally from short sequencing reads. In this study we assess nine different computational tools that have recently been developed to address this challenge. We simulated 100 wastewater sequence samples consisting of SARS-CoV-2 BA.1, BA.2, and Delta lineages, in various mixtures, as well as a Delta-Omicron recombinant and a synthetic 'novel' lineage. Most tools performed well in identifying the true lineages present and estimating their relative abundances and were generally robust to variation in sequencing depth and read length. While many tools identified lineages present down to 1 % frequency, results were more reliable above a 5 % threshold. The presence of an unknown synthetic lineage, which represents an unclassified SARS-CoV-2 lineage, increases the error in relative abundance estimates of other lineages, but the magnitude of this effect was small for most tools. The tools also varied in how they labelled novel synthetic lineages and recombinants. While our simulated dataset represents just one of many possible use cases for these methods, we hope it helps users understand potential sources of error or bias in wastewater sequencing analysis and to appreciate the commonalities and differences across methods.

Latency-associated DNA methylation patterns among HIV-1 infected individuals with distinct disease progression courses or antiretroviral virologic response

DNA methylation is one of the epigenetic modifications that configures gene transcription programs. This study describes the DNA methylation profile of HIV-infected individuals with distinct characteristics related to natural and artificial viremia control. Sheared DNA from circulating mononuclear cells was subjected to target enrichment bisulfite sequencing designed to cover CpG-rich genomic regions. Gene expression was assessed through RNA-seq. Hypermethylation in virologic responders was highly distributed closer to Transcription Start Sites (p-value = 0.03). Hyper and hypomethylation levels within TSS adjacencies varied according to disease progression status (Kruskal–Wallis, p < 0.001), and specific differentially methylated regions associated genes were identified for each group. The lower the promoter methylation, the higher the gene expression in subjects undergoing virologic failure (R = − 0.82, p = 0.00068). Among the inversely correlated genes, those supporting glycolysis and its related pathways were hypomethylated and up-regulated in virologic failures. Disease progression heterogeneity was associated with distinct DNA methylation patterns in terms of rates and distribution. Methylation was associated with the expression of genes sustaining intracellular glucose metabolism in subjects undergoing antiretroviral virologic failure. Our findings highlight that DNA methylation is associated with latency, disease progression, and fundamental cellular processes.

Genomics and epidemiology for gastric adenocarcinomas (GE4GAC): a Brazilian initiative to study gastric cancer

Gastric cancer (GC) is the fifth most common type of cancer worldwide with high incidences in Asia, Central, and South American countries. This patchy distribution means that GC studies are neglected by large research centers from developed countries. The need for further understanding of this complex disease, including the local importance of epidemiological factors and the rich ancestral admixture found in Brazil, stimulated the implementation of the GE4GAC project. GE4GAC aims to embrace epidemiological, clinical, molecular and microbiological data from Brazilian controls and patients with malignant and pre-malignant gastric disease. In this letter, we summarize the main goals of the project, including subject and sample accrual and current findings

Caracterização computacional de neoantígenos e sua implicação prognóstica / Computational characterization of neoantigens and their prognostic implications

Mutações somáticas não sinônimas podem iniciar a tumorigênese e, também, uma resposta citotóxica antitumoral. Com o desenvolvimento das tecnologias de sequenciamento, tornou-se possível identificar as mutações em todos os genes humanos e, consequentemente, as variantes que induzem uma resposta imune (neoantígenos), representando uma oportunidade para pacientes que possam se beneficiar de imunoterapias, mas também um desafio com a necessidade de várias camadas de informações e a integração computacional de vários tipos de dados. Neste trabalho, foi desenvolvido o pipeline de identificação de neoantígeno neo2P, o qual realiza a integração completa de todos os passos necessários para a detecção e neoantígenos e apresentou uma eficiência computacional superior de até seis vezes em comparação com outro método. Além disso, foi proposto um score para priorizaração das mutações somáticas a partir da distribuição dos níveis da expressão gênica de 9.679 pacientes de 32 projetos do TCGA, o qual apresentou um poder de discriminação (AUC) próximo ou superior a 0.7 na maioria das coortes avaliadas. O neo2P foi aplicado em um conjunto de dados de pacientes com melanoma e foram identificados aspectos adicionais da relação de neoantígenos e aspectos imunes, como a expressão de alguns genes marcadores que podem estar relacionados com a resposta ao tratamento. Adicionalmente, a carga de neoantígenos detectados pelo neo2P estratificou, de maneira significativa, pacientes respondedores (R) e não respondedores (NR) quando comparado com o marcador TMB

Somatic non-synonymous mutations can initiate tumorigenesis and, conversely, anti-tumor cytotoxic T cell (CTL) responses. With the development of next-generation sequencing, it has become feasible to detect mutation-derived neoantigens within exome and thereby predict potential neoantigens, which represents an opportunity to patients that may be treated with immunotherapies, but also a challenge due to multiple layers of information and a computational integration of several types of data. In this work, it was developed a neoantigen identification pipeline called neo2P, which integrates all the necessary steps involved for neoantigen detection and presented a six-times superior computational efficiency compared to another method. In addition, a score was proposed to prioritize somatic mutations based on the distribution of gene expression levels in 9,679 patients from 32 TCGA projects, which showed a stratification ability (AUC) close to or greater than 0.7 in most evaluated cohorts.neo2P was applied to a dataset of patients with melanoma and additional aspects of the relationship between neoantigens and immune aspects were identified, such as the expression of some marker genes that may be related to the treatment response. Additionally, the neoantigen load detected by neo2P significantly stratified responders (R) and non-responders (NR) patients when compared to the TMB marker