Systematic rare variant analyses identify RAB32 as a susceptibility gene for familial Parkinson's disease.

Despite substantial progress, causal variants are identified only for a minority of familial Parkinson's disease (PD) cases, leaving high-risk pathogenic variants unidentified1,2. To identify such variants, we uniformly processed exome sequencing data of 2,184 index familial PD cases and 69,775 controls. Exome-wide analyses converged on RAB32 as a novel PD gene identifying c.213C > G/p.S71R as a high-risk variant presenting in ~0.7% of familial PD cases while observed in only 0.004% of controls (odds ratio of 65.5). This variant was confirmed in all cases via Sanger sequencing and segregated with PD in three families. RAB32 encodes a small GTPase known to interact with LRRK2 (refs. 3,4). Functional analyses showed that RAB32 S71R increases LRRK2 kinase activity, as indicated by increased autophosphorylation of LRRK2 S1292. Here our results implicate mutant RAB32 in a key pathological mechanism in PD-LRRK2 kinase activity5-7-and thus provide novel insights into the mechanistic connections between RAB family biology, LRRK2 and PD risk.

Stratified analyses refine association between TLR7 rare variants and severe COVID-19.

Despite extensive global research into genetic predisposition for severe COVID-19, knowledge on the role of rare host genetic variants and their relation to other risk factors remains limited. Here, 52 genes with prior etiological evidence were sequenced in 1,772 severe COVID-19 cases and 5,347 population-based controls from Spain/Italy. Rare deleterious TLR7 variants were present in 2.4% of young (<60 years) cases with no reported clinical risk factors (n=378), compared to 0.24% of controls (odds ratio (OR)=12.3, p=1.27x10-10). Incorporation of the results of either functional assays or protein modeling led to a pronounced increase in effect size (ORmax=46.5, p=1.74x10-15). Association signals for X-chromosomal TLR7 were also detected in the female-only subgroup, suggesting the existence of additional mechanisms beyond X-linked recessive inheritance in males. Additionally, supporting evidence was generated for a contribution to severe COVID-19 of the previously implicated genes IFNAR2, IFIH1 and TBK1. Our results refine the genetic contribution of rare TLR7 variants to severe COVID-19, and strengthen evidence for the etiological relevance of genes in the interferon signaling pathway.

Interleukin-1ß Polymorphisms Are Genetic Markers of Susceptibility to Periprosthetic Joint Infection in Total Hip and Knee Arthroplasty.

Periprosthetic joint infections (PJIs) are serious complications of prosthetic surgery. The criteria for the diagnosis of PJI integrate clinical and laboratory findings in a complex and sometimes inconclusive workflow. Host immune factors hold potential as diagnostic biomarkers in bone and joint infections. We reported that the humoral pattern-recognition molecule long pentraxin 3 (PTX3) predicts PJI in total hip and knee arthroplasty (THA and TKA, respectively). If and how genetic variation in PTX3 and inflammatory genes that affect its expression (IL-1ß, IL-6, IL-10, and IL-17A) contributes to the risk of PJI is unknown. We conducted a case-control study on a Caucasian historic cohort of THA and TKA patients who had prosthesis explant due to PJI (cases) or aseptic complications (controls). Saliva was collected from 93 subjects and used to extract DNA and genotype PTX3, IL-1ß, IL-6, IL-10, and IL-17A single-nucleotide polymorphisms (SNPs). Moreover, the concentration of IL-1ß, IL-10, and IL-6 was measured in synovial fluid and plasma. No association was found between PTX3 polymorphisms and PJI; however, the AGG haplotype, encompassing rs2853550, rs1143634, and rs1143627 in IL-1ß, was linked to the infection (p = 0.017). Also, synovial levels of all inflammatory markers were higher in cases than in controls, and a correlation emerged between synovial concentration of PTX3 and that of IL-1ß in cases only (Spearman r = 0.67, p = 0.004). We identified a relationship between rs2853550 and the synovial concentration of IL-1ß and PTX3. Our findings suggest that IL-1ß SNPs could be used for the early identification of THA and TKA patients with a high risk of infection.

Improving predictive accuracy in primary biliary cholangitis: A new genetic risk score.

BACKGROUND AND AIMS: Genetic variants influence primary biliary cholangitis (PBC) risk. We established and tested an accurate polygenic risk score (PRS) using these variants. METHODS: Data from two Italian cohorts (OldIT 444 cases, 901 controls; NewIT 255 cases, 579 controls) were analysed. The latest international genome-wide meta-analysis provided effect size estimates. The PRS, together with human leukocyte antigen (HLA) status and sex, was included in an integrated risk model. RESULTS: Starting from 46 non-HLA genes, 22 variants were selected. PBC patients in the OldIT cohort showed a higher risk score than controls: -.014 (interquartile range, IQR, -.023, .005) versus -.022 (IQR -.030, -.013) (p < 2.2 × 10-16). For genetic-based prediction, the area under the curve (AUC) was .72; adding sex increased the AUC to .82. Validation in the NewIT cohort confirmed the model's accuracy (.71 without sex, .81 with sex). Individuals in the top group, representing the highest 25%, had a PBC risk approximately 14 times higher than that of the reference group (lowest 25%; p < 10-6). CONCLUSION: The combination of sex and a novel PRS accurately discriminated between PBC cases and controls. The model identified a subset of individuals at increased risk of PBC who might benefit from tailored monitoring.

A genome-wide association study for survival from a multi-centre European study identified variants associated with COVID-19 risk of death.

The clinical manifestations of SARS-CoV-2 infection vary widely among patients, from asymptomatic to life-threatening. Host genetics is one of the factors that contributes to this variability as previously reported by the COVID-19 Host Genetics Initiative (HGI), which identified sixteen loci associated with COVID-19 severity. Herein, we investigated the genetic determinants of COVID-19 mortality, by performing a case-only genome-wide survival analysis, 60 days after infection, of 3904 COVID-19 patients from the GEN-COVID and other European series (EGAS00001005304 study of the COVID-19 HGI). Using imputed genotype data, we carried out a survival analysis using the Cox model adjusted for age, age2, sex, series, time of infection, and the first ten principal components. We observed a genome-wide significant (P-value < 5.0 × 10-8) association of the rs117011822 variant, on chromosome 11, of rs7208524 on chromosome 17, approaching the genome-wide threshold (P-value = 5.19 × 10-8). A total of 113 variants were associated with survival at P-value < 1.0 × 10-5 and most of them regulated the expression of genes involved in immune response (e.g., CD300 and KLR genes), or in lung repair and function (e.g., FGF19 and CDH13). Overall, our results suggest that germline variants may modulate COVID-19 risk of death, possibly through the regulation of gene expression in immune response and lung function pathways.

Low-grade parental gonosomal mosaicism in CHD2 siblings with Smith-Magenis-like syndrome.

Loss-of-function CHD2 (chromodomain helicase DNA-binding protein 2) mutations are associated with a spectrum of neurodevelopmental disorders often including early-onset generalized seizures, photosensitivity, and epileptic encephalopathies. Patients show psychomotor delay/intellectual disability (ID), autistic features, and behavior disorders, such as aggression and impulsivity. Most reported cases are sporadic with description of germline mosaicism only in two families. We detect the first case of parental gonosomal CHD2 mosaicism disclosed by two brothers showing mild ID, born to healthy parents. The eldest brother has a history of drug-controlled generalized tonic-clonic seizures and displays sleep disorder and aggressive behavior suggestive of Smith-Magenis syndrome (SMS). Analysis of brothers' DNAs by next-generation sequencing (NGS) custom gene panel for pediatric epilepsy and/or ID disclosed in both the same pathogenic CHD2 variant. Additional NGS experiment on genomic DNA from parents' peripheral blood and from buccal swab raised the suspicion of low-grade gonosomal mosaicism in the unaffected mother subsequently confirmed by digital polymerase chain reaction (dPCR). This report underlines as worthwhile CHD2 screening in individuals presenting ID/developmental delay, with/without epilepsy, and behavior and sleep disorders suggestive of SMS. Detecting a CHD2 variant should prime testing probands' parents by NGS coupled to dPCR on different tissues to exclude/confirm gonosomal mosaicism and define the recurrence risk.

Congenital fibrinogen disorders: a retrospective clinical and genetic analysis of the Prospective Rare Bleeding Disorders Database.

ABSTRACT: Congenital fibrinogen deficiency (CFD) is a rare bleeding disorder caused by mutations in FGA, FGB, and FGG. We sought to comprehensively characterize patients with CFD using PRO-RBDD (Prospective Rare Bleeding Disorders Database). Clinical phenotypes, laboratory, and genetic features were investigated using retrospective data from the PRO-RBDD. Patients were classified from asymptomatic to grade 3 based on their bleeding severity. In addition, FGA, FGB, and FGG were sequenced to find causative variants. A total of 166 CFD cases from 16 countries were included, of whom 123 (30 afibrinogenemia, 33 hypofibrinogenemia, 55 dysfibrinogenemia, and 5 hypodysfibrinogenemia) were well characterized. Considering the previously established factor activity and antigen level thresholds, bleeding severity was correctly identified in 58% of the cases. The rates of thrombotic events among afibrinogenemic and hypofibrinogenemic patients were relatively similar (11% and 10%, respectively) and surprisingly higher than in dysfibrinogenemic cases. The rate of spontaneous abortions among 68 pregnancies was 31%, including 86% in dysfibrinogenemic women and 14% with hypofibrinogenemia. Eighty-six patients received treatment (69 on-demand and/or 17 on prophylaxis), with fibrinogen concentrates being the most frequently used product. Genetic analysis was available for 91 cases and 41 distinct variants were identified. Hotspot variants (FGG, p.Arg301Cys/His and FGA, p.Arg35Cys/His) were present in 51% of dysfibrinogenemia. Obstetric complications were commonly observed in dysfibrinogenemia. This large multicenter study provided a comprehensive insight into the clinical, laboratory, and genetic history of patients with CFDs. We conclude that bleeding severity grades were in agreement with the established factor activity threshold in nearly half of the cases with quantitative defects.

Male awareness of prostate cancer risk remains poor in relatives of women with germline variants in DNA-repair genes.

Abstract. Objective: The aim of this study is to evaluate male awareness of developing prostate cancer (PCa) in families with germline DNA-repair genes (DRG) variants. Materials and methods: Data were collected from a prospective, monocentric cohort study. The study was conducted in a university hospital with a multidisciplinary approach to the patient (collaboration of the Departments of Oncology, Urology, Pathology, Radiology, and Medical Genetics Laboratory). We recruited healthy males, relatives of families of women with breast or ovarian cancer who tested positive for pathogenic variants (PVs) or likely pathogenic variants (LPVs) in DRGs. A dedicated PCa screening was designed and offered to men aged 35 to 69 years, based on early visits with digital rectal examination (DRE), prostate health index (PHI) measurement, multiparametric magnetic resonance imaging (mpMRI) and, if necessary, targeted/systematic prostate biopsies. The primary endpoint was to evaluate the willingness of healthy men from families with a DRG variants detected in female relatives affected with breast and/or ovarian cancer to be tested for the presence of familial PVs. The secondary endpoints were the acceptance to participate if resulted positive and compliance with the screening programme. Results: Over 1256 families, of which 139 resulted positive for PVs in DRGs, we identified 378 'healthy' men aged between 35 and 69 years old. Two hundred sixty-one (69.0%) refused to be tested for DRG variants, 66 (17.5%) declared to have been previously tested, and 51 (13.5%) males were interested to be tested. Between those previously tested and those who accepted to be tested, 62 (53.0%) were positive for a DRG variant, and all of them accepted to participate in the subsequent surveillance steps. The main limitation is that is a single-centre study and a short follow-up. Conclusions: All men tested positive for a DRG variants agreed to go under the surveillance scheme. However, only 31% of 'men at risk' (i.e., relative of a DRG variant carrier) expressed their willingness to be tested for the familial DRG variant. This observation strongly supports the urgent need to implement awareness of genetic risk for PCa within the male population.

Mechanisms by which the cystic fibrosis transmembrane conductance regulator may influence SARS-CoV-2 infection and COVID-19 disease severity.

Patients with cystic fibrosis (CF) exhibit pronounced respiratory damage and were initially considered among those at highest risk for serious harm from SARS-CoV-2 infection. Numerous clinical studies have subsequently reported that individuals with CF in North America and Europe-while susceptible to severe COVID-19-are often spared from the highest levels of virus-associated mortality. To understand features that might influence COVID-19 among patients with cystic fibrosis, we studied relationships between SARS-CoV-2 and the gene responsible for CF (i.e., the cystic fibrosis transmembrane conductance regulator, CFTR). In contrast to previous reports, we found no association between CFTR carrier status (mutation heterozygosity) and more severe COVID-19 clinical outcomes. We did observe an unexpected trend toward higher mortality among control individuals compared with silent carriers of the common F508del CFTR variant-a finding that will require further study. We next performed experiments to test the influence of homozygous CFTR deficiency on viral propagation and showed that SARS-CoV-2 production in primary airway cells was not altered by the absence of functional CFTR using two independent protocols. On the contrary, experiments performed in vitro strongly indicated that virus proliferation depended on features of the mucosal fluid layer known to be disrupted by absent CFTR in patients with CF, including both low pH and increased viscosity. These results point to the acidic, viscous, and mucus-obstructed airways in patients with cystic fibrosis as unfavorable for the establishment of coronaviral infection. Our findings provide new and important information concerning relationships between the CF clinical phenotype and severity of COVID-19.

Evaluation of Semen Self-Sampling Yield Predictors and CTC Isolation by Multi-Color Flow Cytometry for Liquid Biopsy of Localized Prostate Cancer.

Liquid biopsy (LB) for prostate cancer (PCa) detection could represent an alternative to biopsy. Seminal fluid (SF) is a source of PCa-specific biomarkers, as 40% of ejaculate derives from the prostate. We tested the feasibility of an SF-based LB by evaluating the yield of semen self-sampling in a cohort of >750 patients with clinically localized PCa. The overall SF collection yield was 18.2% (39% when considering only compliant patients), with about a half of the patients (53.15%) not consenting to SF donation. Independent favorable predictors for SF collection were younger age and lower prostate volume. We implemented a protocol to enrich prostate-derived cells by multi-color flow cytometry and applied it on SF and urine samples from 100 patients. The number of prostate-enriched cells (SYTO-16+ PSMA+ CD45-) was variable, with higher numbers of cells isolated from SF than urine (p value < 0.001). Putative cancer cells (EpCAMhigh) were 2% of isolated cells in both specimens. The fraction of EpCAMhigh cells over prostate-enriched cells (PSMA+) significantly correlated with patient age in both semen and urine, but not with other clinical parameters, such as Gleason Score, ISUP, or TNM stage. Hence, enumeration of prostate-derived cells is not sufficient to guide PCa diagnosis; additional molecular analyses to detect patient-specific cancer lesions will be needed.

Genome-wide epistasis study highlights genetic interactions influencing severity of COVID-19.

Coronavirus disease 2019 (COVID-19) caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) may lead to life-threatening respiratory symptoms. Understanding the genetic basis of the prognosis of COVID-19 is important for risk profiling of potentially severe symptoms. Here, we conducted a genome-wide epistasis study of COVID-19 severity in 2243 patients with severe symptoms and 12,612 patients with no or mild symptoms from the UK Biobank, followed by a replication study in an independent Spanish cohort (1416 cases, 4382 controls). Our study highlighted 3 interactions with genome-wide significance in the discovery phase, nominally significant in the replication phase, and enhanced significance in the meta-analysis. For example, the lead interaction was found between rs9792388 upstream of PDGFRL and rs3025892 downstream of SNAP25, where the composite genotype of rs3025892 CT and rs9792388 CA/AA showed higher risk of severe disease than any other genotypes (P = 2.77 × 10-12, proportion of severe cases = 0.24 ~ 0.29 vs. 0.09 ~ 0.18, genotypic OR = 1.96 ~ 2.70). This interaction was replicated in the Spanish cohort (P = 0.002, proportion of severe cases = 0.30 ~ 0.36 vs. 0.14 ~ 0.25, genotypic OR = 1.45 ~ 2.37) and showed enhanced significance in the meta-analysis (P = 4.97 × 10-14). Notably, these interactions indicated a possible molecular mechanism by which SARS-CoV-2 affects the nervous system. The first exhaustive genome-wide screening for epistasis improved our understanding of genetic basis underlying the severity of COVID-19.

Regulation of inflammation and protection against invasive pneumococcal infection by the long pentraxin PTX3.

Streptococcus pneumoniae is a major pathogen in children, elderly subjects, and immunodeficient patients. Pentraxin 3 (PTX3) is a fluid-phase pattern recognition molecule (PRM) involved in resistance to selected microbial agents and in regulation of inflammation. The present study was designed to assess the role of PTX3 in invasive pneumococcal infection. In a murine model of invasive pneumococcal infection, PTX3 was strongly induced in non-hematopoietic (particularly, endothelial) cells. The IL-1ß/MyD88 axis played a major role in regulation of the Ptx3 gene expression. Ptx3-/- mice presented more severe invasive pneumococcal infection. Although high concentrations of PTX3 had opsonic activity in vitro, no evidence of PTX3-enhanced phagocytosis was obtained in vivo. In contrast, Ptx3-deficient mice showed enhanced recruitment of neutrophils and inflammation. Using P-selectin-deficient mice, we found that protection against pneumococcus was dependent upon PTX3-mediated regulation of neutrophil inflammation. In humans, PTX3 gene polymorphisms were associated with invasive pneumococcal infections. Thus, this fluid-phase PRM plays an important role in tuning inflammation and resistance against invasive pneumococcal infection.

Genetic susceptibility to severe COVID-19.

Severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is the etiologic agent of the coronavirus disease 2019 (COVID-19) pandemic. Clinical manifestations of the disease range from an asymptomatic condition to life-threatening events and death, with more severe courses being associated with age, male sex, and comorbidities. Besides these risk factors, intrinsic characteristics of the virus as well as genetic factors of the host are expected to account for COVID-19 clinical heterogeneity. Genetic studies have long been recognized as fundamental to identify biological mechanisms underlying congenital diseases, to pinpoint genes/proteins responsible for the susceptibility to different inherited conditions, to highlight targets of therapeutic relevance, to suggest drug repurposing, and even to clarify causal relationships that make modifiable some environmental risk factors. Though these studies usually take long time to be concluded and, above all, to translate their discoveries to patients' bedside, the scientific community moved really fast to deliver genetic signals underlying different COVID-19 phenotypes. In this Review, besides a concise description of COVID-19 symptomatology and of SARS-CoV-2 mechanism of infection, we aimed to recapitulate the current literature in terms of host genetic factors that specifically associate with an increased severity of the disease.

In-depth genetic and molecular characterization of diaphanous related formin 2 (DIAPH2) and its role in the inner ear.

Diaphanous related formins are regulatory cytoskeletal protein involved in actin elongation and microtubule stabilization. In humans, defects in two of the three diaphanous genes (DIAPH1 and DIAPH3) have been associated with different types of hearing loss. Here, we investigate the role of the third member of the family, DIAPH2, in nonsyndromic hearing loss, prompted by the identification, by exome sequencing, of a predicted pathogenic missense variant in DIAPH2. This variant occurs at a conserved site and segregated with nonsyndromic X-linked hearing loss in an Italian family. Our immunohistochemical studies indicated that the mouse ortholog protein Diaph2 is expressed during development in the cochlea, specifically in the actin-rich stereocilia of the sensory outer hair cells. In-vitro studies showed a functional impairment of the mutant DIAPH2 protein upon RhoA-dependent activation. Finally, Diaph2 knock-out and knock-in mice were generated by CRISPR/Cas9 technology and auditory brainstem response measurements performed at 4, 8 and 14 weeks. However, no hearing impairment was detected. Our findings indicate that DIAPH2 may play a role in the inner ear; further studies are however needed to clarify the contribution of DIAPH2 to deafness.

The circular RNA landscape in multiple sclerosis: Disease-specific associated variants and exon methylation shape circular RNA expression profile.

BACKGROUND: Circular RNAs (circRNAs) are a class of non-coding RNAs increasingly emerging as crucial actors in the pathogenesis of human diseases, including autoimmune and neurological disorders as multiple sclerosis (MS). Despite several efforts, the mechanisms regulating circRNAs expression are still largely unknown and the circRNA profile and regulation in MS-relevant cell models has not been completely investigated. In this work, we aimed at exploring the global landscape of circRNA expression in MS patients, also evaluating a possible correlation with their genetic and epigenetic background. METHODS: We performed RNA-seq experiments on circRNA-enriched samples, derived from peripheral blood mononuclear cells (PBMCs) of 10 MS patients and 10 matched controls and performed differential circRNA expression. The genetic background was evaluated using array genotyping, and an expression quantitative trait loci (eQTL) analysis was carried out. RESULTS: Expression analysis revealed 166 differentially expressed circRNAs in MS patients, 125 of which are downregulated. One of the top dysregulated circRNAs, hsa_circ_0007990, derives from the PGAP3 gene, encoding a protein relevant for the control of autoimmune responses. The downregulation of this circRNA was confirmed in two independent replication cohorts, suggesting its implementation as a possible RNA-based biomarker. The eQTL analysis evidenced a significant association between 89 MS-associated loci and the expression of at least one circRNA, suggesting that MS-associated variants could impact on disease pathogenesis by altering circRNA profiles. Finally, we found a significant correlation between exon methylation and circRNA expression levels, supporting the hypothesis that epigenetic features may play an important role in the definition of the cell circRNA pool. CONCLUSION: We described the circRNA expression profile of PBMCs in MS patients, suggesting that MS-associated variants may tune the expression levels of circRNAs acting as "circ-QTLs", and proposing a role for exon-based DNA methylation in regulating circRNA expression.

Artificial intelligence for precision medicine in autoimmune liver disease.

Autoimmune liver diseases (AiLDs) are rare autoimmune conditions of the liver and the biliary tree with unknown etiology and limited treatment options. AiLDs are inherently characterized by a high degree of complexity, which poses great challenges in understanding their etiopathogenesis, developing novel biomarkers and risk-stratification tools, and, eventually, generating new drugs. Artificial intelligence (AI) is considered one of the best candidates to support researchers and clinicians in making sense of biological complexity. In this review, we offer a primer on AI and machine learning for clinicians, and discuss recent available literature on its applications in medicine and more specifically how it can help to tackle major unmet needs in AiLDs.

GWAS and autoimmunity: What have we learned and what next.

Autoimmune diseases are common conditions characterized by loss of tolerance, female predominance and a remarkable heterogeneity among different populations. Most often they are polygenic and several genetic loci have been linked with the risk of developing autoimmune diseases. However, causal inference is difficult. When the genomic revolution began there were high hopes of translating fast genetic analyses to the bedside but this has proven to be challenging. Nonetheless, over the last decade, fine-mapping strategies have greatly improved; one of the most significant research lines focuses on the in vivo and ex vivo definition of the effect of genetic variants within the target tissues and within specific subpopulations of immune cells that are involved in the disease pathogenesis. This strategy also includes the longitudinal tracking of a large number of immunophenotypes in many individuals to build a large reference atlas for variant characterization. In this review, we discuss the results obtained by GWAS in autoimmune diseases and review recent advances in fine mapping strategies. More importantly, we discuss gaps and future directions.

Genetics of Primary Biliary Cholangitis.

Primary biliary cholangitis (PBC) is a rare disease of the liver characterized by an autoimmune attack on the small bile ducts. PBC is a complex trait, meaning that a large list of genetic factors interacts with environmental agents to determine its onset. Genome-wide association studies have had a huge impact in fostering research in PBC, but many steps need still to be done compared with other autoimmune diseases of similar prevalence. This review presents the state-of-the-art regarding the genetic architecture of PBC and provides some thoughtful reflections about possible future lines of research, which can be helpful to fill the missing heritability gap in PBC.

LLM-PBC: Logic Learning Machine-Based Explainable Rules Accurately Stratify the Genetic Risk of Primary Biliary Cholangitis.

BACKGROUND: The application of Machine Learning (ML) to genetic individual-level data represents a foreseeable advancement for the field, which is still in its infancy. Here, we aimed to evaluate the feasibility and accuracy of an ML-based model for disease risk prediction applied to Primary Biliary Cholangitis (PBC). METHODS: Genome-wide significant variants identified in subjects of European ancestry in the recently released second international meta-analysis of GWAS in PBC were used as input data. Quality-checked, individual genomic data from two Italian cohorts were used. The ML included the following steps: import of genotype and phenotype data, genetic variant selection, supervised classification of PBC by genotype, generation of "if-then" rules for disease prediction by logic learning machine (LLM), and model validation in a different cohort. RESULTS: The training cohort included 1345 individuals: 444 were PBC cases and 901 were healthy controls. After pre-processing, 41,899 variants entered the analysis. Several configurations of parameters related to feature selection were simulated. The best LLM model reached an Accuracy of 71.7%, a Matthews correlation coefficient of 0.29, a Youden's value of 0.21, a Sensitivity of 0.28, a Specificity of 0.93, a Positive Predictive Value of 0.66, and a Negative Predictive Value of 0.72. Thirty-eight rules were generated. The rule with the highest covering (19.14) included the following genes: RIN3, KANSL1, TIMMDC1, TNPO3. The validation cohort included 834 individuals: 255 cases and 579 controls. By applying the ruleset derived in the training cohort, the Area under the Curve of the model was 0.73. CONCLUSIONS: This study represents the first illustration of an ML model applied to common variants associated with PBC. Our approach is computationally feasible, leverages individual-level data to generate intelligible rules, and can be used for disease prediction in at-risk individuals.