Biology subjects

Gecz, J.

Publications and source records attributed to Gecz, J..

6 recordsLinked to original sources

bioRxiv2026

Knowledge Inclusive Machine Learning for Disease Gene Prioritisation

The predictive performance of machine learning models depends on the context available to them. In disease gene prioritisation, this context comprises two forms: specific context from sample-level experimental data, such as gene expression and protein-protein interaction networks, and general context from accumulated and curated biological knowledge capturing established relationships among genes, diseases, and pathways. Neither is sufficient alone: experimental data are sensitive to dataset-specific noise and lack broader biological grounding, while curated knowledge lacks the resolution required for gene-level discrimination. Consequently, most machine learning approaches relying solely on experimental data risk learning spurious correlations rather than underlying biology. Here we introduce Knowledge Inclusive Machine Learning (KIML), a paradigm that integrates both context types within a unified analytical pipeline. KIML combines experimental data with two types of general context: literature-derived representations from PubMed and structured biomedical knowledge graphs. We evaluate the approach on Developmental and Epileptic Encephalopathy and benchmark it against recent methods using publicly available datasets. Performance is assessed using temporal-split evaluation and biological evaluations, including ontology enrichment analysis. KIML consistently outperforms existing approaches, providing improved predictive accuracy and biologically meaningful insights. Furthermore, the framework generates interpretable explanations of gene prioritisation and demonstrates strong generalisability across six additional diseases.

genetics↗

bioRxiv2025

TMT-based quantitative proteomic assessment of Vicia sativa induced neurotoxicity by β-cyano-L-alanine and γ-glutamyl-β-cyano-L-alanine in SH-SY5Y cells.

{beta}-Cyano-L-alanine (BCA) and {gamma}-glutamyl-{beta}-cyano-L-alanine (GBCA) are the primary antinutritional compounds present within the seeds of the high protein and drought tolerant orphan legume Vicia sativa. Evidence of neurotoxicity is limited to symptom analysis from animal feed trials whilst the molecular mechanisms underlying suspected neurotoxicity in monogastric animals are largely unknown. In this study, we first optimised an in vitro cell-based assay for rapid testing of BCA and GBCA toxicity in the retinoic acid differentiated SH-SY5Y human neuroblastoma cells. Using this system, we then performed proteomics analyses to determine dysregulated expression of proteins in BCA or GBCA treated of differentiated SH-SY5Y cells. Our findings indicate that BCA affects expression of proteins involved in DNA damage and translation whilst GBCA treatment causes dysregulation of those involved in mitosis and cell cycle. Following BCA treatment, we identified changes in many proteins previously suggested to have close association with neurodegenerative diseases including amyotrophic lateral sclerosis and Alzheimers disease as well as cancers. Following GBCA treatment, dysregulation of proteins involved in apoptosis pathways was observed. Finally, the lack of common dysregulated proteins and pathways in BCA or GBCA treated cells indicated that they most likely cause neurotoxicity via distinct mechanisms. SignificanceAntinutritional compounds or toxins limit the use of many potential grain crops including V. sativa (Common Vetch), and BCA and GBCA are the two principal toxic compounds in Common Vetch grain. Vetch grain toxicity is associated with neurotoxic symptoms in animals. Characterizing the proteome of BCA- and GBCA-treated neural cells gives insights into the excitotoxicity mechanism and identification of biomarkers for screening higher quality grain. Quantitative tandem mass tag (TMT) mass spectrometry-based proteome profiling of BCA treated neural cells characterized 6,827 proteins, of which 26 were significantly up-regulated and 73 significantly downregulated. However, TMT proteome analysis of GBCA treated cells identified 76 significantly up regulated and 86 downregulated proteins. These results provide insight into the toxicity of BCA and GBCA and enable future identification of putative direct protein targets. We describe for the first time BCA playing a role in modulating the expression DNA damage proteins and translation, and GBCA playing a role in modulating the expression of mitosis and cell cycle proteins, elucidating the mechanisms of plant-derived toxins in mammalian cell neurotoxicity HighlightsO_LIEstablished in vitro cell-based toxicity assay for {beta}-cyano-L-alanine and {gamma}-glutamyl-{beta}-cyano-L-alanine. C_LIO_LI{beta}-Cyano-L-alanine treatment dysregulated many proteins associated with DNA damage in retinoic acid differentiated SH-SY5Y cells. C_LIO_LI{gamma}-Glutamyl-{beta}-cyano-L-alanine treatment dysregulated proteins associated with cell cycle. C_LI

pharmacology and toxicology↗

bioRxiv2024

Optical genome mapping enables accurate repeat expansion testing

Short tandem repeats (STRs) are amongst the most abundant class of variations in human genomes and are meiotically and mitotically unstable which leads to expansions and contractions. STR expansions are frequently associated with genetic disorders, with the size of expansions often correlating with the severity and age of onset. Therefore, being able to accurately detect the total repeat expansion length and to identify potential somatic repeat instability is important. Current standard of care (SOC) diagnostic assays include laborious repeat-primed PCR-based tests as well as Southern blotting, which are unable to precisely determine long repeat expansions and/or require a separate set-up for each locus. Sequencing-based assays have proven their potential for the genome-wide detection of repeat expansions but have not yet replaced these diagnostic assays due to their inaccuracy to detect long repeat expansions (short-read sequencing) and their costs (long-read sequencing). Here, we tested whether optical genome mapping (OGM) can efficiently and accurately identify the STR length and assess the stability of known repeat expansions. We performed OGM for 85 samples with known clinically relevant repeat expansions in DMPK, CNBP and RFC1, causing myotonic dystrophy type 1 and 2 and cerebellar ataxia, neuropathy and vestibular areflexia syndrome (CANVAS), respectively. After performing OGM, we applied three different repeat expansion detection workflows, i.e. manual de novo assembly, local guided assembly (local-GA) and molecule distance script of which the latter two were developed as part of this study. The first two workflows estimated the repeat size for each of the two alleles, while the third workflow was used to detect potential somatic instability. The estimated repeat sizes were compared to the repeat sizes reported after the SOC and concordance between the results was determined. All except one known repeat expansions above the pathogenic repeat size threshold were detected by OGM, and allelic differences were distinguishable, either between wildtype and expanded alleles, or two expanded alleles for recessive cases. An apparent strength of OGM over current SOC methods was the more accurate length measurement, especially for very long repeat expansion alleles, with no upper size limit. In addition, OGM enabled the detection of somatic repeat instability, which was detected in 9/30 DMPK, 23/25 CNBP and 4/30 RFC1 samples, leveraging the analysis of intact, native DNA molecules. In conclusion, for tandem repeat expansions larger than [~]300 bp, OGM provides an efficient method to identify exact repeat lengths and somatic repeat instability with high confidence across multiple loci simultaneously, enabling the potential to provide a significantly improved and generic genome-wide assay for repeat expansion disorders.

genomics↗

bioRxiv2020

A survey of rare epigenetic variation in 23,116 human genomes identifies disease-relevant epivariations and novel CGG expansions

There is growing recognition that epivariations, most often recognized as promoter hypermethylation events that lead to gene silencing, are associated with a number of human diseases. However, little information exists on the prevalence and distribution of rare epigenetic variation in the human population. In order to address this, we performed a survey of methylation profiles from 23,116 individuals using the Illumina 450k array. Using a robust outlier approach, we identified 4,452 unique autosomal epivariations, including potentially inactivating promoter methylation events at 384 genes linked to human disease. For example, we observed promoter hypermethylation of BRCA1 and LDLR at population frequencies of ~1 in 3,000 and ~1 in 6,000 respectively, suggesting that epivariations may underlie a fraction of human disease which would be missed by purely sequence-based approaches. Using expression data, we confirmed that many epivariations are associated with outlier gene expression. Analysis of SNV data and monozygous twin pairs suggests that approximately two thirds of epivariations segregate in the population secondary to underlying sequence mutations, while one third are likely sproradic events that occur post-zygotically. We identified 25 loci where rare hypermethylation coincided with the presence of an unstable CGG tandem repeat, and validated the presence of novel CGG expansions at several of these, identifying the molecular defect underlying most of the known folate-sensitive fragile sites in the genome. Our study provides a catalog of rare epigenetic changes in the human genome, gives insight into the underlying origins and consequences of epivariations, and identifies many novel hypermethylated CGG repeat expansions.

genomics↗

bioRxiv2018

Common genetic variants contribute to risk of rare severe neurodevelopmental disorders

There are thousands of rare human disorders caused by a single deleterious, protein-coding genetic variant 1. However, patients with the same genetic defect can have different clinical presentation 2-4, and some individuals carrying known disease-causing variants can appear unaffected 5. What explains these differences? Here, we show in a cohort of 6,987 children with heterogeneous severe neurodevelopmental disorders expected to be almost entirely monogenic that 7.7% of variance in risk is attributable to inherited common genetic variation. We replicated this genome wide common variant burden by showing that it is over-transmitted from parents to children in an independent sample of 728 trios from the same cohort. Our common variant signal is significantly positively correlated with genetic predisposition to fewer years of schooling, decreased intelligence, and risk of schizophrenia. We found that common variant risk was not significantly different between individuals with and without a known protein-coding diagnostic variant, suggesting that common variant risk is not confined to patients without a monogenic diagnosis. In addition, previously published common variant scores for autism, height, birth weight, and intracranial volume were all correlated with those traits within our cohort, suggesting that phenotypic expression in individuals with monogenic disorders is affected by the same variants as the general population. Our results demonstrate that common genetic variation affects both overall risk and clinical presentation in disorders typically considered to be monogenic.

genetics↗

bioRxiv2017

Abnormal cell sorting underlies the unique X-linked inheritance of PCDH19 Epilepsy

X-linked diseases typically exhibit more severe phenotypes in males than females. In contrast, Protocadherin 19 (PCDH19) mutations cause epilepsy in heterozygous females but spare hemizygous males. The cellular mechanism responsible for this unique pattern of X-linked inheritance is unknown. We show that PCDH19 contributes to highly specific combinatorial adhesion codes such that mosaic expression of Pcdh19 in heterozygous female mice leads to striking sorting between WT PCDH19- and null PCDH19-expressing cells in the developing cortex, correlating with altered network activity. Complete deletion of PCDH19 in heterozygous mice abolishes abnormal cell sorting and restores normal network activity. Furthermore, we identify variable cortical malformations in PCDH19 epilepsy patients. Our results highlight the role of PCDH19 in determining specific adhesion codes during cortical development and how disruption of these codes is associated with the unique X-linked inheritance of PCDH19 epilepsy.

developmental biology↗