Despite the increasing availability of genomic data and enhanced data analysis procedures, predicting the severity of associated diseases remains elusive in the absence of clinical descriptors. To address this challenge, we have focused on the K_V7.2 voltage-gated potassium channel gene (KCNQ2), known for its link to developmental delays and various epilepsies, including self-limited benign familial neonatal epilepsy and epileptic encephalopathy. Genome-wide tools often exhibit a tendency to overestimate deleterious mutations, frequently overlooking tolerated variants, and lack the capacity to discriminate variant severity. This study introduces a novel approach by evaluating multiple machine learning (ML) protocols and descriptors. The combination of genomic information with a novel Variant Frequency Index (VFI) builds a robust foundation for constructing reliable gene-specific ML models. The ensemble model, MLe-KCNQ2, formed through logistic regression, support vector machine, random forest and gradient boosting algorithms, achieves specificity and sensitivity values surpassing 0.95 (AUC-ROC > 0.98). The ensemble MLe-KCNQ2 model also categorizes pathogenic mutations as benign or severe, with an area under the receiver operating characteristic curve (AUC-ROC) above 0.67. This study not only presents a transferable methodology for accurately classifying KCNQ2 missense variants, but also provides valuable insights for clinical counseling and aids in the determination of variant severity. The research context emphasizes the necessity of precise variant classification, especially for genes like KCNQ2, contributing to the broader understanding of gene-specific challenges in the field of genomic research. The MLe-KCNQ2 model stands as a promising tool for enhancing clinical decision making and prognosis in the realm of KCNQ2-related pathologies.

The presence of mycotoxins in food poses a significant risk to food safety, and it is essential to develop effective and safe detoxification strategies. In this study, we demonstrate the strong ability of Pichia fermentans KCB21_L2, a yeast isolated from kefir, to eliminate aflatoxin B₁, fumonisin B₁ and ocratoxin A. Viable cells removed aflatoxin B₁ and fumonisin B₁ more efficiently than heat-inactivated cells, particularly at pH values of 5.5 and 7.0, suggesting the involvement of an active removal process. Subsequently, we evaluated the capacity of P. fermentans KCB21_L2 to remove mycotoxins at high concentrations and investigated the underlying molecular and cellular responses. The yeast effectively eliminated high levels of all three mycotoxins. Transcriptional analysis revealed the activation of metabolic pathways related to amino acid catabolism and fatty acid metabolism, likely reflecting an adaptive stress response. However, no significant upregulation of specific genes related to mycotoxin-degrading enzymes was observed. In conclusion, the reduction process may involve multiple factors, including stress response pathways, possible production of organic acids, adsorption of mycotoxins into the cell wall, and constitutively expressed enzymes capable of degrading mycotoxins. In general, these findings highlight the multifactorial nature of yeast-mediated mycotoxin removal and establish P. fermentans KCB21_L2 as a promising candidate for safe biological decontamination in food systems.

Advances in DNA sequencing technologies have revolutionized rare disease diagnosis, resulting in an increasing volume of available genomic data. Despite this wealth of information and improved procedures to combine data from various sources, identifying the pathogenic causal variants and distinguishing between severe and benign variants remains a key challenge. Mutations in the K _v 7.2 voltage-gated potassium channel gene ( KCNQ2 ) have been linked to different subtypes of epilepsies, such as benign familial neonatal epilepsy (BFNE) and epileptic encephalopathy (EE). To date, there is a wide variety of genome-wide computational tools aiming at predicting the pathogenicity of variants. However, previous reports suggest that these genome-wide tools have limited applicability to the KCNQ2 gene related diseases due to overestimation of deleterious mutations and failure to correctly identify benign variants, being, therefore, of limited use in clinical practice. In this work, we found that combining readily available features, such as AlphaFold structural information, Missense Tolerance Ratio (MTR) and other commonly used protein descriptors, provides foundations to build reliable gene-specific machine learning ensemble models. Here, we present a transferable methodology able to accurately predict the pathogenicity of KCNQ2 missense variants with unprecedented sensitivity and specificity scores above 90%.