Learning with uncertainty to accelerate the discovery of histone lysine-specific demethylase 1A (KDM1A/LSD1) inhibitors.

Machine learning including modern deep learning models has been extensively used in drug design and screening. However, reliable prediction of molecular properties is still challenging when exploring out-of-domain regimes, even for deep neural networks. Therefore, it is important to understand the uncertainty of model predictions, especially when the predictions are used to guide further experiments. In this study, we explored the utility and effectiveness of evidential uncertainty in compound screening. The evidential Graphormer model was proposed for uncertainty-guided discovery of KDM1A/LSD1 inhibitors. The benchmarking results illustrated that (i) Graphormer exhibited comparative predictive power to state-of-the-art models, and (ii) evidential regression enabled well-ranked uncertainty estimates and calibrated predictions. Subsequently, we leveraged time-splitting on the curated KDM1A/LSD1 dataset to simulate out-of-distribution predictions. The retrospective virtual screening showed that the evidential uncertainties helped reduce false positives among the top-acquired compounds and thus enabled higher experimental validation rates. The trained model was then used to virtually screen an independent in-house compound set. The top 50 compounds ranked by two different ranking strategies were experimentally validated, respectively. In general, our study highlighted the importance to understand the uncertainty in prediction, which can be recognized as an interpretable dimension to model predictions.

TransfIGN: A Structure-Based Deep Learning Method for Modeling the Interaction between HLA-A*02:01 and Antigen Peptides.

The intricate interaction between major histocompatibility complexes (MHCs) and antigen peptides with diverse amino acid sequences plays a pivotal role in immune responses and T cell activity. In recent years, deep learning (DL)-based models have emerged as promising tools for accelerating antigen peptide screening. However, most of these models solely rely on one-dimensional amino acid sequences, overlooking crucial information required for the three-dimensional (3-D) space binding process. In this study, we propose TransfIGN, a structure-based DL model that is inspired by our previously developed framework, Interaction Graph Network (IGN), and incorporates sequence information from transformers to predict the interactions between HLA-A*02:01 and antigen peptides. Our model, trained on a comprehensive data set containing 61,816 sequences with 9051 binding affinity labels and 56,848 eluted ligand labels, achieves an area under the curve (AUC) of 0.893 on the binary data set, better than state-of-the-art sequence-based models trained on larger data sets such as NetMHCpan4.1, ANN, and TransPHLA. Furthermore, when evaluated on the IEDB weekly benchmark data sets, our predictions (AUC = 0.816) are better than those of the recommended methods like the IEDB consensus (AUC = 0.795). Notably, the interaction weight matrices generated by our method highlight the strong interactions at specific positions within peptides, emphasizing the model's ability to provide physical interpretability. This capability to unveil binding mechanisms through intricate structural features holds promise for new immunotherapeutic avenues.

A modified CVD method for the synthesis of monolayer MoS2and photoelectric improvement by HfO2passivation.

Molybdenum disulfide (MoS2) is an emerging class of new materials with a wide range of potential practical applications. However, the uncontrollability of monolayer MoS2synthesized by traditional chemical vapor deposition method and the low responsivity of MoS2photodetectors limit its further development in the field of photoelectric detection. To achieve controlled growth of monolayer MoS2and construct MoS2photodetectors with a high responsivity, we propose a novel single crystal growth strategy of high-quality MoS2by controlling the Mo to S vapor ratio near the substrate, and deposit a layer of hafnium oxide (HfO2) on the surface of MoS2to enhance the performance of the pristine metal-semiconductor-metal structure photodetector. At a reverse bias of 8 V, the HfO2passivated MoS2photodetector features an extremely high responsivity of1201AW-1,a response time of around 0.5 s, and a detectivity of7.7×1011Jones.Meanwhile, we deeply investigate the effect of the HfO2layer on the performance of the fabricated MoS2photodetector and propose a physical mechanism to interpret the obtained experiment results. These results might facilitate a better understanding on the performance modulation of the MoS2photodetectors and accelerate the development of MoS2-based optoelectronic devices.

Erastin synergizes with cisplatin via ferroptosis to inhibit ovarian cancer growth in vitro and in vivo.

AIM: Cisplatin-based chemotherapy is the first-line treatment for ovarian cancer. However, acquired resistance to cisplatin treatment or serious side effects often occurs in ovarian cancer, and thus, there is an urgent need for effective and combined therapies to overcome such obstacles. In the present study, we aimed to uncover synergistic effects between erastin and cisplatin (CDDP) in inhibiting ovarian cancer cell growth by inducing ferroptosis in vitro and in vivo. METHODS: We performed a CCK-8 assay to detect cell viability in response to erastin alone or in combination with cisplatin and provided further confirmation by western blotting analysis. Transmission electron microscopy and flow cytometry analysis were used to depict the characteristics of ferroptosis. In addition, an ovarian cancer tumor xenograft was built to verify the effects in vivo. RESULTS: CDDP induced multiple modes of cell death-including ferroptosis in ovarian cancer cell lines. Mechanistically, erastin triggered ferroptosis and increased the levels of reactive oxygen species (ROS) so as to augment the cytotoxic effect of cisplatin. Combination therapy based on CDDP and erastin appeared to maximize the therapeutic effects while minimizing side effects in ovarian cancer both in vitro and in vivo. CONCLUSION: Collectively, our results indicate that erastin works synergistically with cisplatin to inhibit ovarian cancer cell growth, which may be manipulated by a ROS-mediated mechanism that enhances cisplatin therapy, and offers a novel strategy for overcoming cisplatin therapy resistance.

ABCF2, an Nrf2 target gene, contributes to cisplatin resistance in ovarian cancer cells.

Previously, we have demonstrated that NRF2 plays a key role in mediating cisplatin resistance in ovarian cancer. To further explore the mechanism underlying NRF2-dependent cisplatin resistance, we stably overexpressed or knocked down NRF2 in parental and cisplatin-resistant human ovarian cancer cells, respectively. These two pairs of stable cell lines were then subjected to microarray analysis, where we identified 18 putative NRF2 target genes. Among these genes, ABCF2, a cytosolic member of the ABC superfamily of transporters, has previously been reported to contribute to chemoresistance in clear cell ovarian cancer. A detailed analysis on ABCF2 revealed a functional antioxidant response element (ARE) in its promoter region, establishing ABCF2 as an NRF2 target gene. Next, we investigated the contribution of ABCF2 in NRF2-mediated cisplatin resistance using our stable ovarian cancer cell lines. The NRF2-overexpressing cell line, containing high levels of ABCF2, was more resistant to cisplatin-induced apoptosis compared to its control cell line; whereas the NRF2 knockdown cell line with low levels of ABCF2, was more sensitive to cisplatin treatment than its control cell line. Furthermore, transient overexpression of ABCF2 in the parental cells decreased apoptosis and increased cell viability following cisplatin treatment. Conversely, knockdown of ABCF2 using specific siRNA notably increased apoptosis and decreased cell viability in cisplatin-resistant cells treated with cisplatin. This data indicate that the novel NRF2 target gene, ABCF2, plays a critical role in cisplatin resistance in ovarian cancer, and that targeting ABCF2 may be a new strategy to improve chemotherapeutic efficiency.

Primary Fallopian Tube Carcinoma: A Single-Institution Experience of 101 Cases: A Retrospective Study.

OBJECTIVE: This study aimed to identify the prognostic factors for primary fallopian tube carcinoma. METHODS: A retrospective analysis was conducted of the patients treated with primary surgery and adjuvant chemotherapy at the Obstetrics and Gynecology Hospital of Fudan University from February 2003 to December 2010. Cox proportional hazards model was used for univariate and multivariate survival analysis. RESULTS: Included in this study were 101 patients with a median follow-up of 64 months and a mean age of 57 years. Latzko triad symptom of abdominal pain, vaginal bleeding or discharge, and palpable pelvic mass was reported in 14 patients, and elevated CA 125 (≥ 35 U/mL) was found in 63. Four patients were classified as grade 1, 31 were grade 2, and 66 were grade 3. The distribution of International Federation of Gynecology and Obstetrics stage was 33 at stage I, 28 at stage II, 39 at stage III, and 1 at stage IV. Ninety patients underwent optimal tumor debulking in which residual tumor was no larger than 1 cm, and 67 patients received no fewer than 6 cycles of postoperative chemotherapy with paclitaxel and carboplatin (TP)-based regimen. Recurrence occurred in 44 patients after a median of 20 months (range, 1-72 months). The 5-year overall survival rate was 67.7%, and the 5-year disease-free survival was 57.4%. Multivariate analysis revealed that International Federation of Gynecology and Obstetrics stage (I-II) [hazard ratio (HR), 2.670; 95% confidence interval (CI), 1.316-5.418; P = 0.007 vs HR, 2.716; 95% CI, 1.416-5.211; P = 0.003], pelvic lymphadenectomy (HR, 0.274; 95% CI, 0.136-0.555; P < 0.001 vs HR, 0.449; 95% CI, 0.227-0.888; P = 0.021), and cycles (≥ 6) of chemotherapy (HR, 0.480; 95% CI, 0.246-0.937; P = 0.031 vs HR, 0.521; 95% CI, 0.276-0.985; P = 0.045) might serve as independent predictors of both overall survival and disease-free survival. CONCLUSIONS: Preoperative diagnosis of fallopian tube carcinoma is difficult due to the silent course of this neoplasm. Comprehensive surgical staging including pelvic lymphadenectomy followed by adequate cycles of chemotherapy is an important strategy to improve patients' prognosis.

Cytoprotective Role of Heme Oxygenase-1 in Cancer Chemoresistance: Focus on Antioxidant, Antiapoptotic, and Pro-Autophagy Properties.

Chemoresistance remains the foremost challenge in cancer therapy. Targeting reactive oxygen species (ROS) manipulation is a promising strategy in cancer treatment since tumor cells present high levels of intracellular ROS, which makes them more vulnerable to further ROS elevation than normal cells. Nevertheless, dynamic redox evolution and adaptation of tumor cells are capable of counteracting therapy-induced oxidative stress, which leads to chemoresistance. Hence, exploring the cytoprotective mechanisms of tumor cells is urgently needed to overcome chemoresistance. Heme oxygenase-1 (HO-1), a rate-limiting enzyme of heme degradation, acts as a crucial antioxidant defense and cytoprotective molecule in response to cellular stress. Recently, emerging evidence indicated that ROS detoxification and oxidative stress tolerance owing to the antioxidant function of HO-1 contribute to chemoresistance in various cancers. Enhanced HO-1 expression or enzymatic activity was revealed to promote apoptosis resistance and activate protective autophagy, which also involved in the development of chemoresistance. Moreover, inhibition of HO-1 in multiple cancers was identified to reversing chemoresistance or improving chemosensitivity. Here, we summarize the most recent advances regarding the antioxidant, antiapoptotic, and pro-autophagy properties of HO-1 in mediating chemoresistance, highlighting HO-1 as a novel target for overcoming chemoresistance and improving the prognosis of cancer patients.

Kinome-wide polypharmacology profiling of small molecules by multi-task graph isomorphism network approach.

Prediction of the interactions between small molecules and their targets play important roles in various applications of drug development, such as lead discovery, drug repurposing and elucidation of potential drug side effects. Therefore, a variety of machine learning-based models have been developed to predict these interactions. In this study, a model called auxiliary multi-task graph isomorphism network with uncertainty weighting (AMGU) was developed to predict the inhibitory activities of small molecules against 204 different kinases based on the multi-task Graph Isomorphism Network (MT-GIN) with the auxiliary learning and uncertainty weighting strategy. The calculation results illustrate that the AMGU model outperformed the descriptor-based models and state-of-the-art graph neural networks (GNN) models on the internal test set. Furthermore, it also exhibited much better performance on two external test sets, suggesting that the AMGU model has enhanced generalizability due to its great transfer learning capacity. Then, a naïve model-agnostic interpretable method for GNN called edges masking was devised to explain the underlying predictive mechanisms, and the consistency of the interpretability results for 5 typical epidermal growth factor receptor (EGFR) inhibitors with their structure‒activity relationships could be observed. Finally, a free online web server called KIP was developed to predict the kinome-wide polypharmacology effects of small molecules (http://cadd.zju.edu.cn/kip).

Integrated Analysis of Ferroptosis-Related Biomarker Signatures to Improve the Diagnosis and Prognosis Prediction of Ovarian Cancer.

Ovarian cancer remains the most lethal gynecological malignancy. Ferroptosis, a specialized form of iron-dependent, nonapoptotic cell death, plays a crucial role in various cancers. However, the contribution of ferroptosis to ovarian cancer is poorly understood. Here, we characterized the diagnostic, prognostic, and therapeutic value of ferroptosis-related genes in ovarian cancer by analyzing transcriptomic data from The Cancer Genome Atlas and Gene Expression Omnibus databases. A reliable 10-gene ferroptosis signature (HIC1, ACSF2, MUC1, etc.) for the diagnosis of ovarian cancer was identified. Notably, we constructed and validated a novel prognostic signature including three FRGs: HIC1, LPCAT3, and DUOX1. We also further developed a risk score model based on these three genes which divided ovarian cancer patients into two risk groups. Functional analysis revealed that immune response and immune-related pathways were enriched in the high-risk group. Meanwhile, the tumor microenvironment was distinct between the two groups, with more M2 Macrophage infiltration and higher expression of key immune checkpoint molecules in the high-risk group than in the other group. Low-risk patients exhibited more favorable immunotherapy and chemotherapy responses. We conclude that crosstalk between ferroptosis and immunity may contribute to the worse prognosis of patients in the high-risk group. In particular, HIC1 showed both diagnostic and prognostic value in ovarian cancer. In vitro experiments demonstrated that inhibition of HIC1 improved drug sensitivity of chemotherapy and immunotherapy agents by inducing ferroptosis. Our findings provide new insights into the potential role of FRGs in the early detection, prognostic prediction, and individualized treatment decision-making for ovarian cancer patients.

Unexpected primary fallopian tube carcinoma during gynecological operations: Clinicopathological and prognostic factors analyses of 67 cases.

OBJECTIVE: The aim of the study was to analyze the clinicopathologic features, the survival rate, and the prognostic factors of women with unexpected primary fallopian tube carcinoma diagnosed during gynecological operations. MATERIALS AND METHODS: We reviewed medical records of patients with unexpected primary fallopian tube carcinoma at the Obstetrics and Gynecology Hospital of Fudan University between January 2004 to December 2017. The survival analysis was based on the Kaplan-Meier method, and the results were compared using the log-rank test. Cox regression analysis was used to determine factors affecting survival. RESULTS: Sixty-seven patients with unexpected primary fallopian tube carcinoma were identified. The 5-year overall survival was 49.7%, the mean overall survival was 64 months [95% confidence interval (CI) 54-74], and the median overall survival was 59 months (95% CI 49-69). The mean follow-up time was 53.9 months (range 5-141 months). The most common clinical presentation was adnexal mass (38.8%), followed by vaginal bleeding (16.4%) and no specific symptom (13.4%). Cytoreductive surgery was performed initially in 57 (85.1%) patients. Residual disease was optimal in 56 (83.6%) patients and suboptimal in 11 (16.4%) patients. The histological subtype was predominantly the serous type (88.1%). 44 patients (65.7%) were diagnosed at Stage I/II postoperatively. 23 (34.3%) patients were in Stage III/IV. 51 patients (76.1%) had gone through laparoscopic surgery, 16 patients (23.9%) were performed laparotomy. Univariate analyses on overall survival revealed that only the International Federation of Gynecology and Obstetrics (FIGO)stage [p < 0.001; Hazard Ratio (HR), 6.433; 95% CI, 2.274-18.199], residual tumor (p = 0.014; HR, 4.957; 95% CI, 1.378-17.831) were significant prognostic factor. Pelvic lymphadenectomy did not show association with overall survival in our univariate or multivariate analyses. After an observation period of 70 months, we found an increased overall survival in the group of without lymphadenectomy. CONCLUSIONS: The diagnosis of primary fallopian tube carcinoma is rarely considered preoperatively. The early stage and optimal debulking surgery with residual tumor ≤1 cm are important independent factors to improve patients' prognosis. However, there were no statistically significant correlations between lymphadenectomy and prognosis. The value of lymph node sampling or dissection needs to be reconsidered.

Nrf2 induces cisplatin resistance via suppressing the iron export related gene SLC40A1 in ovarian cancer cells.

Induction of Nuclear factor erythroid 2 (NF-E2)-related factor 2 (Nrf2) has been demonstrated to be involved in cisplatin resistance in ovarian cancer. Solute carrier family 40 member 1 (SLC40A1) is an iron exporter, which possesses many putative Nrf2 binding sites. Here we hypothesize that it may be a possible downstream gene of Nrf2. Elevated level of Nrf2 and reduced level of SLC40A1 were found in cisplatin-resistant ovarian cancer cells as compared with cisplatin-sensitive ovarian cancer cells. Exogenous knockdown of Nrf2 leaded to increased expression of SLC40A1. While overexpression of Nrf2 resulted in decreased expression of SLC40A1. Chromatin immunoprecipitation (ChIP) and dual-luciferase reporter assay revealed that Nrf2 inhibited the transcription of SLC40A1. Overexpression of SLC40A1 was able to reverse cisplatin resistance induced by Nrf2, while knockdown of SLC40A1 restored cisplatin resistance and increased iron concentration. Desferal, an iron chelator, was found to overcome cisplatin resistance through iron deprivation. Its function was boosted when combined with brusatol, an Nrf2 inhibitor. Taken together, this study first demonstrated that Nrf2 could transcriptionally suppress the expression of SLC40A1. Iron overload induced by SLC40A1 resulted in cisplatin resistance in ovarian cancer. Targeting iron metabolism may be a new therapeutic strategy to reverse drug resistance in ovarian cancer treatment.

Retrospective analysis of the clinicopathologic and prognostic characteristics of stage I placental site trophoblastic tumor in China.

OBJECTIVE: To investigate clinicopathologic features and identify prognostic factors of placental site trophoblastic tumor (PSTT). METHODS: In a retrospective study, data were analyzed from patients with stage I PSTT treated at a tertiary hospital in Shanghai, China, from January 2007 to May 2013. Univariate log-rank tests were used to examine the association between clinicopathologic characteristics and overall survival and disease-free survival (DFS). RESULTS: In total, seven patients had stage I PSTT. Mean age was 31.6 years (range 22-42). Four patients had term delivery as the outcome of their antecedent pregnancy. Six had a ß-human chorionic gonadotropin (ß-hCG) serum concentration of less than 10 000 mIU/mL. Among five patients who underwent hysterectomy combined with chemotherapy, one had recurrent disease. One patient received fertility-preserving therapy and achieved complete remission. The mean 5-year overall survival and DFS were 100% and 86%, respectively. Maximum ß-HCG concentration of at least 10 000 mIU/mL and a mitotic index of more than 5 mitotic counts per 10 high-power fields were associated with disease recurrence (both P=0.014). CONCLUSION: Pretreatment ß-hCG concentration and mitotic index might be predictors of recurrence among patients with PSTT. Fertility-preserving therapy might be practical in some patients.

Induction of autophagy contributes to cisplatin resistance in human ovarian cancer cells.

Cisplatin resistance is a major challenge in the clinical treatment of ovarian cancer, of which the underlying mechanisms remain unknown. The aim of the present study was to explore the role of autophagy in cisplatin resistance in ovarian cancer cells. A2780cp cisplatin-resistant ovarian carcinoma cells and the A2780 parental cell line, were used as a model throughout the present study. The cell viability was determined using a water soluble tetrazolium salt-8 assay, and western blot analysis was performed to determine the protein expression levels of microtubule-associated protein 1 light chain 3 (LC3 I and LC3 II), and Beclin 1. Beclin 1 small interfering (si)RNA and 3-methyladenine (3-MA) were used to determine whether inhibition of autophagy may re-sensitize cisplatin-resistant cells to cisplatin. The ultrastructural analysis of autophagosomes was performed using transmission electron microscopy, and apoptosis was measured by flow cytometry. In both A2780cp and A2780 cells, cisplatin induced the formation of autophagosomes and upregulated the expression levels of autophagy protein markers, LC3 II and Beclin 1. However, the levels of autophagy were significantly higher in A2780cp cells, as compared with the A2780 cells. The combined treatment of cisplatin with 3-MA, the autophagy pharmacological inhibitor, increased the cell death rate, but had no effects on apoptosis, as compared with cisplatin treatment alone in A2780cp cells. However, inhibition of autophagy by siRNA knockdown of Beclin 1 expression enhanced cisplatin-induced cell death and apoptosis. The findings of the present study suggest that autophagy has a protective role in human ovarian cancer cells, and that targeting autophagy may promote chemotherapeutic sensitivity.

Nrf2 induces cisplatin resistance through activation of autophagy in ovarian carcinoma.

UNLABELLED: Cisplatin resistance is a major problem affecting ovarian carcinoma treatment. NF-E2-related factor 2 (Nrf2), a nuclear transcription factor, plays an important role in chemotherapy resistance. However, the underlying mechanism by which Nrf2 mediates cisplatin chemoresistance is unclear. METHODS: The human ovarian carcinoma cell line, A2780, and its cisplatin-resistant variant, A2780cp were cultivated. Cell viability was determined with WST-8 assay. Western blot was applied to detect the expression of Nrf2, Nrf2 target genes, and autophagy-related proteins. RNA interference was used to knock down target genes. Annexin V and propidium iodide (PI) staining was utilized to quantify apoptosis. The ultrastructural analysis of autophagosomes was performed by transmission electron microscopy (TEM). RESULTS: Nrf2 and its targeting genes, NQO1 and HO-1, are overexpressed in A2780cp cells compared with A2780 cells. Knocking down Nrf2 sensitized A2780cp cells to cisplatin treatment and decreased autophagy-related genes, Atg3, Atg6, Atg12 and p62 in both mRNA and protein levels. Furthermore, we demonstrated that in both cell lines cisplatin could induce the formation of autophagosomes and upregulate the expression of autophagy-related genes Atg3, Atg6 and Atg12. Treatment with an autophagy inhibitor, 3-Methyladenine (3-MA), or beclin 1 siRNA enhanced cisplatin-induced cell death in A2780cp cells, suggesting that inhibition of autophagy renders resistant cells to be more sensitive to cisplatin. Taken together, Nrf2 signaling may regulate cisplatin resistance by activating autophagy. CONCLUSIONS: Nrf2-activated autophagy may function as a novel mechanism causing cisplatin-resistance.