Search

Assessing the mutagenicity of chemicals is an essential task in the drug development process. Usually, databases and other structured sources for AMES mutagenicity exist, which have been carefully and laboriously curated from scientific publications. As knowledge accumulates over time, updating these databases is always an overhead and impractical. In this paper, we first propose the problem of predicting the mutagenicity of chemicals from textual information in scientific publications. More simply, given a chemical and evidence in the natural language form from publications where the mutagenicity of the chemical is described, the goal of the model/algorithm is to predict if it is potentially mutagenic or not. For this, we first construct a golden standard data set and then propose MutaPredBERT, a prediction model fine-tuned on BioLinkBERT based on a question-answering formulation of the problem. We leverage transfer learning and use the help of large transformer-based models to achieve a Macro F1 score of >0.88 even with relatively small data for fine-tuning. Our work establishes the utility of large language models for the construction of structured sources of knowledge bases directly from scientific publications. [ABSTRACT FROM AUTHOR]

Staphylococcus aureus (SA) plays a significant role in numerous serious life-threatening infections that present a major challenge to public health in controlling it, especially those resistant to methicillin (MR) known as (MRSA). These pathogens have resistance to other classes of antimicrobial agents including aminoglycoside molecules which mostly resist it through three types of medically significant enzymes: APH (3')-III, ANT (4')-I and (6')/APH(2"). In this paper, seventy-two MRAS were isolated from different lesions from Al-Diwaniya teaching hospital, and Maternity and Children teaching hospital, during the period from January to July 2018. Disc diffusion method, minimum inhibitory concentration(MIC), and bactericidal (MBC) were carried out to MRAS strains subjected to phenotype and genotype identification as well as to detect AMEs genes. Susceptibility of 29 drugs of MRAS strains was: 100% susceptible to vancomycin and chloramphenicol, but 100% resistant to penicillin, cefoxitin, ceftriaxone, aztreonam, and nitrofurantoin. Thus, a (72.2%) of MRAS were found to be either MDR or XDR including 20 aminoglycoside resistant (AR) strains. Multiple antibiotic resistance (MAR) index of a total of 20(100%)AR strains were recorded high values>0.2 ranged(0.48-0.83) from the maximum MAR index:1. Moreover, MIC and MBC values of vancomycin, for S. aureus, ranged from 2 μ g/mL t o 8 μ g/mL. P henotype r esistance o f M RSA strains to aminoglycoside molecules was: kanamycin 20(27.8%)tobramycin 18(25%); gentamicin 16(22.2%);amikacin 14(19.4%); and netilmicin 8(11.1%). PCR analysis led to all 100% MRSA caring for the mecA gene. Frequency of genes encoding aminoglycosides resistance aac(6')/aph (2");80%, aph (3')-IIIa;45%, and ant (4')-Ia; 35%. The aac(6')/aph (2") and ant (4')-Ia genes was the only determinant of resistance in 5 and 1 strains respectively. Correlation between MRSA-AR strains and AMEs genes was 90%. In conclusion, MRSA strains harbouring the mecA gene are currently widely distributed in the Al-Diwaniya governorate. Co-production with AMEs may increase the risk of the spreading of multiple drug resistance clinical strains in communities and hospitals. [ABSTRACT FROM AUTHOR]