HormoNet: a deep learning approach for hormone-drug interaction prediction.

Several experimental evidences have shown that the human endogenous hormones can interact with drugs in many ways and affect drug efficacy. The hormone drug interactions (HDI) are essential for drug treatment and precision medicine; therefore, it is essential to understand the hormone-drug associations. Here, we present HormoNet to predict the HDI pairs and their risk level by integrating features derived from hormone and drug target proteins. To the best of our knowledge, this is one of the first attempts to employ deep learning approach for prediction of HDI prediction. Amino acid composition and pseudo amino acid composition were applied to represent target information using 30 physicochemical and conformational properties of the proteins. To handle the imbalance problem in the data, we applied synthetic minority over-sampling technique technique. Additionally, we constructed novel datasets for HDI prediction and the risk level of their interaction. HormoNet achieved high performance on our constructed hormone-drug benchmark datasets. The results provide insights into the understanding of the relationship between hormone and a drug, and indicate the potential benefit of reducing risk levels of interactions in designing more effective therapies for patients in drug treatments. Our benchmark datasets and the source codes for HormoNet are available in: https://github.com/EmamiNeda/HormoNet .

Serum and follicular fluid levels of soluble receptor for advanced glycation end-products in women with and without polycystic ovary syndrome.

BACKGROUND: Advanced glycation end products (AGEs) are known to associate with the pathogenesis of several chronic diseases via interaction with their corresponding receptor (RAGE). The soluble forms of RAGE (sRAGE) are considered as anti-inflammatory agents by inhibiting the consequent adverse effects of AGE. We aimed at comparing sRAGE levels in the follicular fluid (FF) and serum of women with or without Polycystic Ovary Syndrome (PCOS) who underwent controlled ovarian stimulation for in vitro fertilisation (IVF). METHODS: A total of forty-five eligible women (26 non-PCOS (control) and 19 patients with PCOS (case)) were included the study. sRAGEs in FF and blood serum were measured using ELISA kit. RESULTS: No statistically significant differences were found in FF and serum sRAGE between case and control groups. Correlation analysis showed a significant and positive relationship between serum levels of sRAGE and FF sRAGE in PCOS (r = 0.639; p = 0.004), in control participants (r = 0.481; p = 0.017), and in total participants (r = 0.552; p = 0.000). Data revealed a statistically significant difference in FF sRAGE concentration among all participants by body mass index (BMI) categories (p = 0.01) and in controls (p = 0.022). Significant differences were found for all the nutrients and AGEs consumption according to Food Frequency Questionnaire in both groups (p = 0.0001). A significant reverse relationship was found between FF levels of sRAGE and AGE in PCOS (r = -0.513; p = 0.025). The concentration of sRAGE in serum and FF is the same in PCOS and control. CONCLUSION: The present study revealed for the first time that there are no statistically significant differences between the concentration of serum sRAGE and FF sRAGE among Iranian women with and without PCOS. However, BMI and dietary intake of AGEs have more significant effects on sRAGE concentration in Iranian women. Future studies in developed and developing countries with larger sample sizes are required to determine the long-term consequences of chronic AGE over consumption and the optimal strategies for minimizing AGE-related pathology, specifically in low income and developing countries.

Fatty Acids in Subcutaneous Adipose Tissue of Pregnant Women with and without Polycystic Ovary Syndrome Are Associated with Genes Related to Steroidogenesis: A Case-Control Study.

BACKGROUND: The qualitative analysis of adipose tissue (AT) is an exciting area for research and clinical applications in several diseases and it is emerging along with the quantitative approach to research on overweight and obese people. While the importance of steroid metabolism in women with polycystic ovary syndrome (PCOS) has been reported, limited data exists on the effective roles of AT in pregnant women suffering from PCOS. The aim of this study was to determine association of fatty acid (FA) profiles with expression of 14 steroid genes in abdominal subcutaneous AT of PCOS vs. non-PCOS pregnant women. MATERIALS AND METHODS: In this case-control study, the AT samples of 36 non-PCOS pregnant women and 12 pregnant women with PCOS (3:1 ratio control: case) who underwent cesarean section were collected. Relationship of expressing gene targets and different features were performed using Pearson correlation analysis on the R 3.6.2 software. The ggplot2 package in R tool was used to draw the plots. RESULTS: Age (31.4 and 31.5 years, P=0.99), body mass index (BMI) (prior pregnancy 26 and 26.5 kg.m-2, P=0.62) and at delivery day (30.1 and 31, P=0.94), gestational period (264 and 267 days, P=0.70) and parity (1.4 and 1.4, P=0.42) of non-PCOS and PCOS pregnant women were similar. Expression of steroidogenic acute regulator (STAR) and 11ß-Hydroxysteroid dehydrogenase (11BHSD2) in non-PCOS pregnant women showed the highest association with eicosapentaenoic acid (EPA, C20:5 n-3, r=0.59, P=0.001) and (r=0.66, P=0.001), respectively. In the all participants, STAR mRNA level showed the greatest association with the EPA fatty acid concentration (P=0.001, r=0.51). CONCLUSION: Our results showed a link between the genes involved in steroid metabolism and fatty acids in AT of pregnant women, especially for omega-3 FA and the gene involved in the first step of steroidogenesis in subcutaneous AT. These findings warrant further studies.

Differences in expression of genes related to steroidgenesis in abdominal subcutaneous adipose tissue of pregnant women with and without PCOS; a case control study.

BACKGROUND: It was reported that steroid-related gene expressions in the adipose tissue (AT) of women differ between women affected with polycystic ovary syndrome (PCOS) and non-PCOS. Although association between PCOS in mother and offspring's health is a crucial issue, there are few studies focusing on AT of pregnant women suffering from PCOS. Our objectives were to determine the differences between mRNA expression levels of key steroid-converting enzymes in abdominal subcutaneous AT of pregnant women afflicted with PCOS and non-PCOS. METHODS: Twelve pregnant women with PCOS (case) and thirty six non-PCOS pregnant women (control) (1:3 ratio; age- and BMI-matched) undergoing cesarean section were enrolled for the present study. Expressions of fifteen genes related to steriodogenesis in abdominal subcutaneous AT were investigated using quantitative real-time PCR. RESULTS: No significant differences were detected with respect to age, BMI (prior pregnancy and at delivery day), gestational period and parity among pregnant women with PCOS and non-PCOS. Most of the sex steroid-converting genes except 17ß-Hydroxysteroid dehydrogenases2 (17BHSD2), were highly expressed on the day of delivery in subcutaneous AT. Women with PCOS showed significantly higher mRNA levels of steroidgenic acute regulator (STAR; P < 0.001), cytochrome P450 monooxygenase (CYP11A1; P < 0.05), 17α-hydroxylase (CYP17A1; P < 0.05), and 11ß-Hydroxysteroid dehydrogenase (11BHSD1 and 11BHSD2; P < 0.05). The expression of steroid 21-hydroxylase (CYP21) in non-PCOS was fourfold higher than those of women with PCOS (P < 0.001). There were no significant differences between relative expression of aromatase cytochrome P450 (CYP19A1), 3ß-hydroxysteroid dehydrogenase (3BHSD1 and 3BHSD2), and 17BHSD family (1, 3, 5, 7, and 12) between the two groups. CONCLUSION: The expression levels of genes related to sex steroids metabolism were similar to age-matched and BMI- matched pregnant non-PCOS and pregnant women with PCOS at delivery day. However, the alterations in gene expressions involved in glucocorticoids and mineralocorticoids metabolism were shown. It is necessary to point out that further studies regarding functional activity are required. More attention should be given to AT of pregnant women with PCOS that was previously ignored.

AptaNet as a deep learning approach for aptamer-protein interaction prediction.

Aptamers are short oligonucleotides (DNA/RNA) or peptide molecules that can selectively bind to their specific targets with high specificity and affinity. As a powerful new class of amino acid ligands, aptamers have high potentials in biosensing, therapeutic, and diagnostic fields. Here, we present AptaNet-a new deep neural network-to predict the aptamer-protein interaction pairs by integrating features derived from both aptamers and the target proteins. Aptamers were encoded by using two different strategies, including k-mer and reverse complement k-mer frequency. Amino acid composition (AAC) and pseudo amino acid composition (PseAAC) were applied to represent target information using 24 physicochemical and conformational properties of the proteins. To handle the imbalance problem in the data, we applied a neighborhood cleaning algorithm. The predictor was constructed based on a deep neural network, and optimal features were selected using the random forest algorithm. As a result, 99.79% accuracy was achieved for the training dataset, and 91.38% accuracy was obtained for the testing dataset. AptaNet achieved high performance on our constructed aptamer-protein benchmark dataset. The results indicate that AptaNet can help identify novel aptamer-protein interacting pairs and build more-efficient insights into the relationship between aptamers and proteins. Our benchmark dataset and the source codes for AptaNet are available in: https://github.com/nedaemami/AptaNet .

Computational predictive approaches for interaction and structure of aptamers.

Aptamers are short single-strand sequences that can bind to their specific targets with high affinity and specificity. Usually, aptamers are selected experimentally via systematic evolution of ligands by exponential enrichment (SELEX), an evolutionary process that consists of multiple cycles of selection and amplification. The SELEX process is expensive, time-consuming, and its success rates are relatively low. To overcome these difficulties, in recent years, several computational techniques have been developed in aptamer sciences that bring together different disciplines and branches of technologies. In this paper, a complementary review on computational predictive approaches of the aptamer has been organized. Generally, the computational prediction approaches of aptamer have been proposed to carry out in two main categories: interaction-based prediction and structure-based predictions. Furthermore, the available software packages and toolkits in this scope were reviewed. The aim of describing computational methods and tools in aptamer science is that aptamer scientists might take advantage of these computational techniques to develop more accurate and more sensitive aptamers.

Differences in fatty acid profiles and desaturation indices of abdominal subcutaneous adipose tissue between pregnant women with and without PCOS.

The objective was to determine the differences in fatty acid (FA) profiles in subcutaneous adipose tissue (AT) between pregnant women with polycystic ovary syndrome (PCOS) and those without PCOS. FA profiles of AT samples from 13 PCOS and 32 non-PCOS, all of whom underwent caesarean section were compared using gas chromatography. Age and BMI in the two groups were similar. Twenty-one FAs were detected and the total saturated FA percentage of experimental groups was similar. While the total monounsaturated FA (MUFA) (p < 0.0004) and desaturase index (18:1 cis-9/18:0; p < 0.03) were higher in PCOS women than non-PCOS women, total polyunsaturated FA (PUFA) was lower in PCOS than non-PCOS women (p < 0.004). Docosahexaenoic acid level of the two groups was similar while α-linolenic acid and eicosapentaenoic acid levels were significantly (p < 0.05) lower in PCOS. Total trans-FA, C18:1 t9 and C18:2t were lower in PCOS women (p < 0.05). These results indicate differences in desaturase index, MUFA and PUFA, especially n-3 FA in AT between age and BMI-matched pregnant PCOS and non-PCOS pregnant subjects. Further studies are warranted to replicate these findings and to investigate potential changes in these profiles in non-pregnant PCOS women.