Dataset Information

Deep generative modeling of the human proteome reveals over a hundred novel genes involved in rare genetic disorders.

ABSTRACT: Identifying causal mutations accelerates genetic disease diagnosis, and therapeutic development. Missense variants present a bottleneck in genetic diagnoses as their effects are less straightforward than truncations or nonsense mutations. While computational prediction methods are increasingly successful at prediction for variants in known disease genes, they do not generalize well to other genes as the scores are not calibrated across the proteome. To address this, we developed a deep generative model, popEVE, that combines evolutionary information with population sequence data and achieves state-of-the-art performance at ranking variants by severity to distinguish patients with severe developmental disorders from potentially healthy individuals. popEVE identifies 442 genes in a cohort of developmental disorder cases, including evidence of 119 novel genetic disorders without the need for gene-level enrichment and without overestimating the prevalence of pathogenic variants in the population. By placing variants on a unified scale, our model offers a comprehensive perspective on the distribution of fitness effects across the entire proteome and the broader human population. popEVE provides compelling evidence for genetic diagnoses even in exceptionally rare single-patient disorders where conventional techniques relying on repeated observations may not be applicable. Interactive web viewer and downloads available at pop.evemodel.org.

SUBMITTER: Orenbuch R

PROVIDER: S-EPMC10705666 | biostudies-literature | 2023 Nov

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Proteome-wide model for human disease genetics.

Orenbuch Rose R Shearer Courtney A CA Kollasch Aaron W AW Spinner Hansen D HD Hopf Thomas A TA van Niekerk Lood L Franceschi Dinko D Dias Mafalda M Frazer Jonathan J Marks Debora S DS

medRxiv : the preprint server for health sciences 20250314

Identifying variants driving disease accelerates both genetic diagnosis and therapeutic development, but missense variants still present a bottleneck as their effects are less straightforward than truncations or nonsense mutations. While computational prediction methods are sufficiently accurate to be of clinical value for variants in <i>known</i> disease genes, they do not generalize well to other genes as the scores are not calibrated across the proteome <sup>1-6</sup> . To address this, we de ...[more]

PMID: 38076790

Similar Datasets

Project description:Intrinsically disordered proteins have dynamic structures through which they play key biological roles. The elucidation of their conformational ensembles is a challenging problem requiring an integrated use of computational and experimental methods. Molecular simulations are a valuable computational strategy for constructing structural ensembles of disordered proteins but are highly resource-intensive. Recently, machine learning approaches based on deep generative models that learn from simulation data have emerged as an efficient alternative for generating structural ensembles. However, such methods currently suffer from limited transferability when modeling sequences and conformations absent in the training data. Here, we develop a novel generative model that achieves high levels of transferability for intrinsically disordered protein ensembles. The approach, named idpSAM, is a latent diffusion model based on transformer neural networks. It combines an autoencoder to learn a representation of protein geometry and a diffusion model to sample novel conformations in the encoded space. IdpSAM was trained on a large dataset of simulations of disordered protein regions performed with the ABSINTH implicit solvent model. Thanks to the expressiveness of its neural networks and its training stability, idpSAM faithfully captures 3D structural ensembles of test sequences with no similarity in the training set. Our study also demonstrates the potential for generating full conformational ensembles from datasets with limited sampling and underscores the importance of training set size for generalization. We believe that idpSAM represents a significant progress in transferable protein ensemble modeling through machine learning.Author summaryProteins are essential molecules in living organisms and some of them have highly dynamical structures, which makes understanding their biological roles challenging. Disordered proteins can be studied through a combination of computer simulations and experiments. Computer simulations are often resource-intensive. Recently, machine learning has been used to make this process more efficient. The strategy is to learn from previous simulations to model the heterogenous conformations of proteins. However, such methods still suffer from poor transferability, meaning that they tend to make incorrect predictions on proteins not seen in training data. In this study, we present idpSAM, a method based on generative artificial intelligence for modeling the structures of disordered proteins. The model was trained using a vast dataset and, thanks to its architecture and training procedure, it performs well on not just proteins in the training set but achieves high levels transferability to proteins unseen in training. This advancement is a step forward in modeling biologically relevant disordered proteins. It shows how the combination of generative modeling and large training sets and can aid us understand how dynamical proteins behave.

Project description:BackgroundPrimary spontaneous pneumomediastinum (PSPM) is a benign condition, but it can be difficult to discriminate from Boerhaave syndrome. The diagnostic difficulty is attributable to a shared constellation of history, signs, and symptoms combined with a poor understanding of the basic vital signs, labs, and diagnostic findings characterizing PSPM. These challenges likely contribute to high resource utilization for diagnosis and management of a benign process.MethodsPatients aged 18 years or older with PSPM were identified from our radiology department's database. A retrospective chart review was performed.ResultsExactly 100 patients with PSPM were identified between March 2001 and November 2019. Demographics and histories correlated well with prior studies: mean age (25 years); male predominance (70%); association with cough (34%), asthma (27%), retching or emesis (24%), tobacco abuse (11%), and physical activity (11%); acute chest pain (75%), and dyspnea (57%) as the first and second most frequent symptoms and subcutaneous emphysema (33%) as the most common sign. We provide the first robust data on presenting vital signs and laboratory values of PSPM, showing that tachycardia (31%) and leukocytosis (30%) were common. No pleural effusion was found in the 66 patients who underwent computed tomography (CT) of the chest. We provide the first data on inter-hospital transfer rates (27%). 79% of transfers were due to concern for esophageal perforation. Most patients were admitted (57%), with an average length of stay (LOS) of 2.3 days, and 25% received antibiotics.ConclusionsPSPM patients frequently present in their twenties with chest pain, subcutaneous emphysema, tachycardia, and leukocytosis. Approximately 25% have a history of retching or emesis and it is this population that must be discriminated from those with Boerhaave syndrome. An esophagram is rarely indicated and observation alone is appropriate in patients under age 40 with a known precipitating event or risk factors for PSPM (e.g., asthma, smoking) if they have no history of retching or emesis. Fever, pleural effusion, and age over 40 are rare in PSPM and should raise concern for esophageal perforation in a patient with a history of retching, emesis, or both.

Dataset Information

Deep generative modeling of the human proteome reveals over a hundred novel genes involved in rare genetic disorders.

Publications

Proteome-wide model for human disease genetics.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets