The identification of optimal drug candidates is very important in drug discovery. Researchers in biology and computational sciences have sought to use machine learning (ML) to efficiently predict drug–target interactions (DTIs). In recent years, according to the emerging usefulness of pretrained models in natural language process (NLPs), pretrained models are being developed for chemical compounds and target proteins. This study sought to improve DTI predictive models using a Bidirectional Encoder Representations from the Transformers (BERT)-pretrained model, ChemBERTa, for chemical compounds. Pretraining features the use of a simplified molecular-input line-entry system (SMILES). We also employ the pretrained ProBERT for target proteins (pretraining employed the amino acid sequences). The BIOSNAP, DAVIS, and BindingDB databases (DBs) were used (alone or together) for learning. The final model, taught by both ChemBERTa and ProtBert and the integrated DBs, afforded the best DTI predictive performance to date based on the receiver operating characteristic area under the curve (AUC) and precision-recall-AUC values compared with previous models. The performance of the final model was verified using a specific case study on 13 pairs of subtrates and the metabolic enzyme cytochrome P450 (CYP). The final model afforded excellent DTI prediction. As the real-world interactions between drugs and target proteins are expected to exhibit specific patterns, pretraining with ChemBERTa and ProtBert could teach such patterns. Learning the patterns of such interactions would enhance DTI accuracy if learning employs large, well-balanced datasets that cover all relationships between drugs and target proteins.

Machine learning techniques are extensively employed in drug discovery, with a significant focus on developing QSAR models that interpret the structural information of potential drugs. In this study, the pre-trained natural language processing (NLP) model, ChemBERTa, was utilized in the drug discovery process. We proposed and evaluated four core model architectures as follows: deep neural network (DNN), encoder, concatenation (concat), and pipe. The DNN model processes physicochemical properties as input, while the encoder model leverages the simplified molecular input line entry system (SMILES) along with NLP techniques. The latter two models, concat and pipe, incorporate both SMILES and physicochemical properties, operating in parallel and with sequential manners, respectively. We collected 5238 entries from DrugBank, including their physicochemical properties and absorption, distribution, metabolism, excretion, and toxicity (ADMET) features. The models’ performance was assessed by the area under the receiver operating characteristic curve (AUROC), with the DNN, encoder, concat, and pipe models achieved 62.4%, 76.0%, 74.9%, and 68.2%, respectively. In a separate test with 84 experimental microsomal stability datasets, the AUROC scores for external data were 78% for DNN, 44% for the encoder, and 50% for concat, indicating that the DNN model had superior predictive capabilities for new data. This suggests that models based on structural information may require further optimization or alternative tokenization strategies. The application of natural language processing techniques to pharmaceutical challenges has demonstrated promising results, highlighting the need for more extensive data to enhance model generalization.

Expression of pro-inflammatory cytokines is influenced by single nucleotide polymorphisms (SNPs) in the promoter regions of the pro-inflammatory cytokine genes, and cytokines are associated with the occurrence of post-chemotherapy cognitive impairment. Hence, the aim of this study was to evaluate the associations between two common pro-inflammatory cytokine gene polymorphisms namely, IL6-174 (rs1800795 G>C) and TNF-308 (rs1800629 G>A), and chemotherapy-associated cognitive impairment (CACI) among Asian early-stage breast cancer patients. In addition, the differential effect of these SNPs on plasma IL-6 and TNF-α levels, and the associations of plasma IL-6 and TNF-α levels with CACI were also assessed. Asian early-stage breast cancer patients (Stage I to III) receiving chemotherapy were prospectively recruited from two cancer centers in Singapore. Patients' cognitive function was longitudinally assessed using the validated FACT-Cog (ver. 3) and an objective computerized battery, Headminder™ at three-time points. Plasma IL-6 and TNF-α levels were analyzed using the multiplex immunoassay, and genotyping was performed using Sanger sequencing. Regression analyses and generalized estimating equation were utilized for statistical analysis. A total of 125 patients were included (mean age: 50.3; Chinese: 80.8%; post-menopausal: 48.0%; 68.0% received anthracycline-based chemotherapy). 36.8% patients experienced self-perceived cognitive impairment, detected in memory (32.8%) and attention (34.2%) domains. Patients with higher levels of anxiety (p<0.001) and insomnia (p = 0.003) also reported more self-perceived cognitive impairment. Higher plasma concentrations of IL-6 were associated with greater severity of self-perceived cognitive impairment (p = 0.001). Polymorphisms of cytokine genes were not associated with expression of plasma cytokines. Present findings further contribute to the growing evidence that supports the role of the pro-inflammatory cytokine IL-6 in the occurrence of cognitive impairment post-chemotherapy. However, genetic polymorphism of these cytokines did not play a major role to the cytokine fluctuations as well as cognitive impairment in this cohort. With an increasing evidence to support the cytokine hypothesis, future studies should investigate the role of anti-inflammatory agents in mitigating the cognitive impairment associated with chemotherapy.

Preliminary evidence suggests that changes in plasma brain-derived neurotrophic factor (BDNF) levels may contribute to the occurrence of chemotherapy-associated cognitive impairment (CACI), and a previous study suggested that carriers of the BDNF Met homozygous genotype are protected from CACI. This multicenter, prospective cohort study involved chemotherapy-receiving early-stage breast cancer (ESBC) patients. Self-perceived cognitive function was longitudinally assessed using the validated FACT-Cog (ver. 3) across three time points: Prior to chemotherapy (T1), during chemotherapy (T2), and at the end of chemotherapy (T3). Plasma BDNF levels were quantified using enzyme-linked immunosorbent assay. Genotyping was performed using Sanger Sequencing. A total of 51 chemotherapy-receiving ESBC patients (mean age: 52.6 ± 9.5 years) were recruited, and 11 patients (21.6%) reported subjective cognitive impairment post-chemotherapy. Overall, there was a reduction in median plasma BDNF levels over time (T1: 5423.0 pg/ml; T2: 5313.6 pg/ml; T3: 4050.3 pg/ml; p < 0.01). After adjusting for confounding factors, longitudinal analysis revealed that BDNF levels were associated with self-reported concentration deficit (p = 0.032). Carriers of Val/Val (p = 0.011) and Val/Met (p = 0.003) BDNF genotypes demonstrated a significant reduction in plasma BDNF levels over time; however, plasma BDNF levels were similar across all time points among Met homozygous carriers (p = 0.107). There was a statistically significant change in BDNF levels post-chemotherapy in ESBC patients, and plasma BDNF levels were associated with self-perceived concentration deficit in patients receiving chemotherapy.