Search

Machine learning (ML) algorithms can handle complex genomic data and identify predictive patterns that may not be apparent through traditional statistical methods. They become popular tools for medical applications including prediction, diagnosis or treatment of complex diseases like rheumatoid arthritis (RA). RA is an autoimmune disease in which genetic factors play a major role. Among the most important genetic factors predisposing to the development of this disease and serving as genetic markers are HLA-DRB and non-HLA genes single nucleotide polymorphisms (SNPs). Another marker of RA is the presence of anticitrullinated peptide antibodies (ACPA) which is correlated with severity of RA. We use genetic data of SNPs in four non-HLA genes (PTPN22, STAT4, TRAF1, CD40 and PADI4) to predict the occurrence of ACPA positive RA in the Polish population. This work is a comprehensive comparative analysis, wherein we assess and juxtapose various ML classifiers. Our evaluation encompasses a range of models, including logistic regression, k-nearest neighbors, naïve Bayes, decision tree, boosted trees, multilayer perceptron, and support vector machines. The top-performing models demonstrated closely matched levels of accuracy, each distinguished by its particular strengths. Among these, we highly recommend the use of a decision tree as the foremost choice, given its exceptional performance and interpretability. The sensitivity and specificity of the ML models is about 70% that are satisfying. In addition, we introduce a novel feature importance estimation method characterized by its transparent interpretability and global optimality. This method allows us to thoroughly explore all conceivable combinations of polymorphisms, enabling us to pinpoint those possessing the highest predictive power. Taken together, these findings suggest that non-HLA SNPs allow to determine the group of individuals more prone to develop RA rheumatoid arthritis and further implement more precise preventive approach.

Rheumatoid arthritis (RA) is an autoimmune disease characterized by chronic inflammation affecting up to 2.0% of adults around the world. The molecular background of RA has not yet been fully elucidated, but RA is classified as a disease in which the genetic background is one of the most significant risk factors. One hallmark of RA is impaired DNA repair observed in patient-derived peripheral blood mononuclear cells (PBMCs). The aim of this study was to correlate the phenotype defined as the efficiency of DNA double-strand break (DSB) repair with the genotype limited to a single-nucleotide polymorphism (SNP) of DSB repair genes. We also analyzed the expression level of key DSB repair genes. The study population contained 45 RA patients and 45 healthy controls. We used a comet assay to study DSB repair after in vitro exposure to bleomycin in PBMCs from patients with rheumatoid arthritis. TaqMan SNP Genotyping Assays were used to determine the distribution of SNPs and the Taq Man gene expression assay was used to assess the RNA expression of DSB repair-related genes. PBMCs from patients with RA had significantly lower bleomycin-induced DNA lesion repair efficiency and we identified more subjects with inefficient DNA repair in RA compared with the control (84.5% vs. 24.4%; OR 41.4, 95% CI, 4.8–355.01). Furthermore, SNPs located within the RAD50 gene (rs1801321 and rs1801320) increased the OR to 53.5 (95% CI, 4.7–613.21) while rs963917 and rs3784099 (RAD51B) to 73.4 (95% CI, 5.3–1011.05). These results were confirmed by decision tree (DT) analysis (accuracy 0.84; precision 0.87, and specificity 0.86). We also found elevated expression of RAD51B, BRCA1, and BRCA2 in PBMCs isolated from RA patients. The findings indicated that impaired DSB repair in RA may be related to genetic variations in DSB repair genes as well as their expression levels. However, the mechanism of this relation, and whether it is direct or indirect, needs to be elucidated.

The increasingly expanding genomic databases generate the need for new tools for their processing and further use. In the paper, a bioinformatics tool, which is a search engine of microsatellite elements—trinucleotide repeat sequences (TRS) in files of FASTA type—is presented. An innovative approach was applied in the tool, which consists of connecting—within one search engine—both mapping of TRS motifs and extracting sequences that are found between the mapped TRS motifs. Accordingly, we present hereby the tool called TRS-omix, which comprises a new engine for searching information on genomes and enables generation of sets of sequences and their number, providing the basis for making comparisons between genomes. In our paper, we showed one of the possibilities of using the software. Using TRS-omix and other IT tools, we showed that we were able to extract sets of DNA sequences that can be assigned only to the genomes of the extraintestinal pathogenic Escherichia coli strains or to the genomes of the intestinal pathogenic Escherichia coli strains, as well as providing the basis for differentiation of the genomes/strains belonging to each of these clinically essential pathotypes.

The branch of informatics that deals with construction and operation of computers built of DNA, is one of the research directions which investigates issues related to the use of DNA as hardware and software. This concept assumes the use of DNA computers due to their biological origin mainly for intelligent, personalized and targeted diagnostics frequently related to therapy. Important elements of this concept are (1) the retrieval of unique DNA sequences using machine learning methods and, based on the results of this process, (2) the construction/design of smart diagnostic biochip projects. The authors of this paper propose a new concept of designing diagnostic biochips, the key elements of which are machine-learning methods and the concept of biomolecular queue automata. This approach enables the scheduling of computational tasks at the molecular level by sequential events of cutting and ligating DNA molecules. We also summarize current challenges and perspectives of biomolecular computer application and machine-learning approaches using DNA sequence data mining.

The Binary State Speciation and Extinction (BiSSE) model is a branching process based model that allows the diversification rates to be controlled by a binary trait. We develop a general approach, based on the BiSSE model, for predicting pathogenicity in bacterial populations from microsatellites profiling data. A comprehensive approach for predicting pathogenicity in E. coli populations is proposed using the state-dependent branching process model combined with microsatellites TRS-PCR profiling. Additionally, we have evaluated the possibility of using the BiSSE model for estimating parameters from genetic data. We analyzed a real dataset (from 251 E. coli strains) and confirmed previous biological observations demonstrating a prevalence of some virulence traits in specific bacterial sub-groups. The method may be used to predict pathogenicity of other bacterial taxa.

Abstract The development of conventional, silicon-based computers has several limitations, including some related to the Heisenberg uncertainty principle and the von Neumann “bottleneck”. Biomolecular computers based on DNA and proteins are largely free of these disadvantages and, along with quantum computers, are reasonable alternatives to their conventional counterparts in some applications. The idea of a DNA computer proposed by Ehud Shapiro’s group at the Weizmann Institute of Science was developed using one restriction enzyme as hardware and DNA fragments (the transition molecules) as software and input/output signals. This computer represented a two-state two-symbol finite automaton that was subsequently extended by using two restriction enzymes. In this paper, we propose the idea of a multistate biomolecular computer with multiple commercially available restriction enzymes as hardware. Additionally, an algorithmic method for the construction of transition molecules in the DNA computer based on the use of multiple restriction enzymes is presented. We use this method to construct multistate, biomolecular, nondeterministic finite automata with four commercially available restriction enzymes as hardware. We also describe an experimental applicaton of this theoretical model to a biomolecular finite automaton made of four endonucleases.