Investigating the Molecular Mechanisms Behind Uncharacterized Cysteine Losses from Prediction of Their Oxidation State

Self Cite

Next generation sequencing technologies are providing increasing amounts of sequencing data, paving the way for improvements in clinical genetics and precision medicine. The interpretation of the observed genomic variants in the light of their phenotypic effects is thus emerging as a crucial task to solve in order to advance our understanding of how exomic variants affect proteins and how the proteins’ functional changes affect human health. Since the experimental evaluation of the effects of every observed variant is unfeasible, Bioinformatics methods are being developed to address this challenge in-silico, by predicting the impact of millions of variants, thus providing insight into the deleteriousness landscape of entire proteomes. Here we show the feasibility of this approach by using the recently developed DEOGEN2 variant-effect predictor to perform the largest in-silico mutagenesis scan to date. We computed the deleteriousness score of 170 million variants over 15000 human proteins and we analysed the results, investigating how the predicted deleteriousness landscape of the proteins relates to known functionally and structurally relevant protein regions and biophysical properties. Moreover, we qualitatively validated our results by comparing them with two mutagenesis studies targeting two specific proteins, showing the consistency of DEOGEN2 predictions with respect to experimental data.

Section: Resultssupporting

confidence: 57%

Section: Resultsmentioning

confidence: 99%

Large-scale in-silico statistical mutagenesis analysis sheds light on the deleteriousness landscape of the human proteome

Raimondi

Orlando

Tabaro³

et al. 2018

Self Cite

“…The second experiment we carried out relates to the discrimination between oxidised and reduced cysteines. Cysteines are an atypical amino acid because the thiol group in its sidechain can undergo oxidation 11 to form a disulfide bond with the thiol group of another cysteine in the same (intra-chain disulfide bonds) or in another protein (inter-chain disulfide bonds). These strong covalent bonds can have structural, allosteric and catalytic roles in proteins 11 .…”

Section: Resultsmentioning

confidence: 99%

“…Cysteines are an atypical amino acid because the thiol group in its sidechain can undergo oxidation 11 to form a disulfide bond with the thiol group of another cysteine in the same (intra-chain disulfide bonds) or in another protein (inter-chain disulfide bonds). These strong covalent bonds can have structural, allosteric and catalytic roles in proteins 11 . The goal of the cysteine oxidation prediction task (OXCYS) is to train a classifier able to discriminate between cysteines that are present in the sequence in their reduced form with respect to cysteines forming a disulfide bond with another cysteine belonging to the same protein chain and it is thus formalized as a binary classification problem.…”

Section: Resultsmentioning

confidence: 99%

“…For this reason, finding the optimal feature representation for each problem is a central issue for bioinformaticians, and improvements in both ML models and the features used have contributed to a great deal of positive results in the field, such as the prediction of protein Secondary Structure 1–3 , Relative Solvent Accessibility 4,5 , Contact Maps 6,7 , homology 8,9 , cysteine oxidation states 10,11 , early folding events 12 , the deleteriousness of variants 13–15 and many more protein-related aspects.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Exploring the limitations of biophysical propensity scales coupled with machine learning for protein sequence analysis

Raimondi

Orlando²,

Vranken

et al. 2019

Self Cite

Machine learning (ML) is ubiquitous in bioinformatics, due to its versatility. One of the most crucial aspects to consider while training a ML model is to carefully select the optimal feature encoding for the problem at hand. Biophysical propensity scales are widely adopted in structural bioinformatics because they describe amino acids properties that are intuitively relevant for many structural and functional aspects of proteins, and are thus commonly used as input features for ML methods. In this paper we reproduce three classical structural bioinformatics prediction tasks to investigate the main assumptions about the use of propensity scales as input features for ML methods. We investigate their usefulness with different randomization experiments and we show that their effectiveness varies among the ML methods used and the tasks. We show that while linear methods are more dependent on the feature encoding, the specific biophysical meaning of the features is less relevant for non-linear methods. Moreover, we show that even among linear ML methods, the simpler one-hot encoding can surprisingly outperform the “biologically meaningful” scales. We also show that feature selection performed with non-linear ML methods may not be able to distinguish between randomized and “real” propensity scales by properly prioritizing to the latter. Finally, we show that learning problem-specific embeddings could be a simple, assumptions-free and optimal way to perform feature learning/engineering for structural bioinformatics tasks.

The correlation between CRB1 variants and the clinical severity of Brazilian patients with different inherited retinal dystrophy phenotypes

Motta

Salles

Costa

et al. 2017

Inherited retinal dystrophies are characterized by progressive retina degeneration and mutations in at least 250 genes have been associated as disease-causing. CRB1 is one of many genes analyzed in molecular diagnosis for inherited retinal dystrophy. Crumbs homolog-1 protein encoded by CRB1 is important for cell-to-cell contact, polarization of epithelial cells and the morphogenesis of photoreceptors. Pathogenic variants in CRB1 lead to a huge variety of phenotypes ranging from milder forms of inherited retinal dystrophy, such as retinitis pigmentosa to more severe phenotypes such as Leber congenital amaurosis. In this study, seven novel likely-pathogenic variants were identified: four missense variants (p.Leu479Pro, p.Ala921Pro, p.Cys948Arg and p.Asp1031Asn), two frameshift deletions (c.2536_2542del7 and c.3460_3461delTG) and one frameshift indel variant (c.276_294delinsTGAACACTGTAC). Furthermore, two patients with cone-rod dystrophy due to mutations in CRB1 were reported, supporting previous data, in which mutations in CRB1 can also cause cone-rod dystrophy. Finally, our data suggested there was a direct relation between phenotype severity and the mutation effect on protein functionality in 15 Brazilian CRB1 patients.