Ligand-based and structure-based studies to develop predictive models for SARS-CoV-2 main protease inhibitors through the 3d-qsar.com portal

The main protease (M^pro) of SARS-Cov-2 is the essential enzyme for maturation of functional proteins implicated in viral replication and transcription. The peculiarity of its specific cleavage site joint with its high degree of conservation among all coronaviruses promote it as an attractive target to develop broad-spectrum inhibitors, with high selectivity and tolerable safety profile. Herein is reported a combination of three-dimensional quantitative structure–activity relationships (3-D QSAR) and comparative molecular binding energy (COMBINE) analysis to build robust and predictive ligand-based and structure-based statistical models, respectively. Models were trained on experimental binding poses of co-crystallized M^pro-inhibitors and validated on available literature data. By means of deep optimization both models’ goodness and robustness reached final statistical values of r²/q² values of 0.97/0.79 and 0.93/0.79 for the 3-D QSAR and COMBINE approaches respectively, and an overall predictiveness values of 0.68 and 0.57 for the SDEP_PRED and AAEP metrics after application to a test set of 60 compounds covered by the training set applicability domain. Despite the different nature (ligand-based and structure-based) of the employed methods, their outcome fully converged. Furthermore, joint ligand- and structure-based structure–activity relationships were found in good agreement with nirmatrelvir chemical features properties, a novel oral M^pro-inhibitor that has recently received U.S. FDA emergency use authorization (EUA) for the oral treatment of mild-to-moderate COVID-19 infected patients. The obtained results will guide future rational design and/or virtual screening campaigns with the aim of discovering new potential anti-coronavirus lead candidates, minimizing both time and financial resources. Moreover, as most of calculation were performed through the well-established web portal 3d-qsar.com the results confirm the portal as a useful tool for drug design.

Using machine learning to predict heavy drinking during outpatient alcohol treatment

BACKGROUND

Accurate clinical prediction supports the effective treatment of alcohol use disorder (AUD) and other psychiatric disorders. Traditional statistical techniques have identified patient characteristics associated with treatment outcomes. However, less work has focused on systematically leveraging these associations to create optimal predictive models. The current study demonstrates how machine learning can be used to predict clinical outcomes in people completing outpatient AUD treatment.

METHOD

We used data from the COMBINE multisite clinical trial (n = 1383) to develop and test predictive models. We identified three priority prediction targets, including (1) heavy drinking during the first month of treatment, (2) heavy drinking during the last month of treatment, and (3) heavy drinking between weekly/bi-weekly sessions. Models were generated using the random forest algorithm. We used "leave sites out" partitioning to externally validate the models in trial sites that were not included in the model training. Stratified model development was used to test for sex differences in the relative importance of predictive features.

RESULTS

Models predicting heavy alcohol use during the first and last months of treatment showed internal cross-validation area under the curve (AUC) scores ranging from 0.67 to 0.74. AUC was comparable in the external validation using data from held-out sites (AUC range = 0.69 to 0.72). The model predicting between-session heavy drinking showed strong classification accuracy in internal cross-validation (AUC = 0.89) and external test samples (AUC range = 0.80 to 0.87). Stratified analyses showed substantial sex differences in optimal feature sets.

CONCLUSION

Machine learning techniques can predict alcohol treatment outcomes using routinely collected clinical data. This technique has the potential to greatly improve clinical prediction accuracy without requiring expensive or invasive assessment methods. More research is needed to understand how best to deploy these models.

Differences in Sociodemographic and Alcohol-Related Clinical Characteristics Between Treatment Seekers and Nontreatment Seekers and Their Role in Predicting Outcomes in the COMBINE Study for Alcohol Use Disorder

BACKGROUND

One of the challenges in early-stage clinical research aimed at developing novel treatments for alcohol use disorder (AUD) is that the enrolled participants are heavy drinkers, but do not seek treatment for AUD.

AIMS

To compare nontreatment seekers with alcohol dependence (AD) from 4 human laboratory studies conducted at Brown University (N = 240; 65.4% male) to treatment seekers with AD from the multisite COMBINE study (N = 1,383; 69.1% male) across sociodemographic and alcohol-related clinical variables and to evaluate whether the variables that significantly differentiate the 2 samples predict the 3 main COMBINE clinical outcomes: time to relapse, percent days abstinent (PDA), and good clinical outcome.

METHODS

Sample characteristics were assessed by parametric and nonparametric testing. Three regression models measured the association between the differing variables and the 3 main COMBINE clinical outcomes.

RESULTS

The nontreatment seekers, compared to the treatment seekers, were more ethnically diverse, less educated, single, and working part-time or unemployed (p's < 0.05); they met fewer DSM-IV AD criteria and had significantly lower scores on alcohol-related scales (p's < 0.05); they were less likely to have a father with alcohol problems (p < 0.0001) and had a significantly earlier age of onset and longer duration of AD (p's < 0.05); they also had significantly more total drinks, drinks per drinking day, heavy drinking days (HDD), and lower PDA in the 30 days prior to baseline (p's < 0.0001 to <0.05). Having more HDD in the 30 days prior to baseline predicted all of the 3 COMBINE clinical outcomes. All the other characteristics mentioned above that differed significantly between the 2 groups predicted at least 1 of the 3 COMBINE clinical outcomes, except for level of education, age of onset, and duration of AD.

CONCLUSIONS

The observed differences between groups should be considered in efforts across participant recruitment at different stages of the development of new treatments for AUD.

Can Alcohol Use Disorder Recovery Include Some Heavy Drinking? A Replication and Extension up to 9 Years Following Treatment

BACKGROUND

Recent research indicates some individuals who engage in heavy drinking following treatment for alcohol use disorder fare as well as those who abstain with respect to psychosocial functioning, employment, life satisfaction, and mental health. The current study evaluated whether these findings replicated in an independent sample and examined associations between recovery profiles and functioning up to 6 years later.

METHODS

Data were from the 3-year and 7- to 9-year follow-ups of subsamples initially recruited for the COMBINE study (3-year follow-up: n = 694; 30.1% female, 21.0% non-White; 7- to 9-year follow-up: n = 127; 38.9% female, 27.8% non-White). Recovery at 3 years was defined by latent profile analyses including measures of health functioning, quality of life, employment, alcohol consumption, and cannabis and other drug use. Functioning at the 7- to 9-year follow-up was assessed using single items of self-rated general health, hospitalizations, and alcohol consumption.

RESULTS

We identified 4 profiles at the 3-year follow-up: (i) low-functioning frequent heavy drinkers (13.9%), (ii) low-functioning infrequent heavy drinkers (15.8%), (iii) high-functioning heavy drinkers (19.4%), and (iv) high-functioning infrequent drinkers (50.9%). At the 7- to 9-year follow-up, the 2 high-functioning profiles had the best self-rated health, and the high-functioning heavy drinking profile had significantly fewer hospitalizations than the low-functioning frequent heavy drinking profile.

CONCLUSIONS

Previous findings showing heterogeneity in recovery outcomes were replicated. Most treatment recipients functioned well for years after treatment, and a subset who achieved stable recovery engaged in heavy drinking and reported good health outcomes up to 9 years after treatment. Results question the long-standing emphasis on drinking practices as a primary outcome, as well as abstinence as a recovery criterion in epidemiologic and treatment outcome research and among stakeholder groups and funding/regulatory agencies. Findings support an expanded recovery research agenda that considers drinking patterns, health, life satisfaction, and functioning.

The first 10 years of the international coordination network for standards in systems and synthetic biology (COMBINE)

This paper presents a report on outcomes of the 10th Computational Modeling in Biology Network (COMBINE) meeting that was held in Heidelberg, Germany, in July of 2019. The annual event brings together researchers, biocurators and software engineers to present recent results and discuss future work in the area of standards for systems and synthetic biology. The COMBINE initiative coordinates the development of various community standards and formats for computational models in the life sciences. Over the past 10 years, COMBINE has brought together standard communities that have further developed and harmonized their standards for better interoperability of models and data. COMBINE 2019 was co-located with a stakeholder workshop of the European EU-STANDS4PM initiative that aims at harmonized data and model standardization for in silico models in the field of personalized medicine, as well as with the FAIRDOM PALs meeting to discuss findable, accessible, interoperable and reusable (FAIR) data sharing. This report briefly describes the work discussed in invited and contributed talks as well as during breakout sessions. It also highlights recent advancements in data, model, and annotation standardization efforts. Finally, this report concludes with some challenges and opportunities that this community will face during the next 10 years.

Active learning strategies with COMBINE analysis: new tricks for an old dog

The COMBINE method was designed to study congeneric series of compounds including structural information of ligand-protein complexes. Although very successful, the method has not received the same level of attention than other alternatives to study Quantitative Structure Active Relationships (QSAR) mainly because lack of ways to measure the uncertainty of the predictions and the need for large datasets. Active learning, a semi-supervised learning approach that makes use of uncertainty to enhance models' performance while reducing the size of the training sets, has been used in this work to address both problems. We propose two estimators of uncertainty: the pool of regressors and the distance to the training set. The performance of the methods has been evaluated by testing the resulting active learning workflows in 3 diverse datasets: HIV-1 protease inhibitors, Taxol-derivatives and BRD4 inhibitors. The proposed strategies were successful in 80% of the cases for the taxol-derivatives and BRD4 inhibitors, while outperformed random selection in the case of the HIV-1 protease inhibitors time-split. Our results suggest that AL-COMBINE might be an effective way of producing consistently superior QSAR models with a limited number of samples.

Specifications of Standards in Systems and Synthetic Biology: Status and Developments in 2017

Standards are essential to the advancement of Systems and Synthetic Biology. COMBINE provides a formal body and a centralised platform to help develop and disseminate relevant standards and related resources. The regular special issue of the Journal of Integrative Bioinformatics aims to support the exchange, distribution and archiving of these standards by providing unified, easily citable access. This paper provides an overview of existing COMBINE standards and presents developments of the last year.

Disruptor of telomeric silencing 1-like (DOT1L): disclosing a new class of non-nucleoside inhibitors by means of ligand-based and structure-based approaches

Chemical inhibition of chromatin-mediated signaling involved proteins is an established strategy to drive expression networks and alter disease progression. Protein methyltransferases are among the most studied proteins in epigenetics and, in particular, disruptor of telomeric silencing 1-like (DOT1L) lysine methyltransferase plays a key role in MLL-rearranged acute leukemia Selective inhibition of DOT1L is an established attractive strategy to breakdown aberrant H3K79 methylation and thus overexpression of leukemia genes, and leukemogenesis. Although numerous DOT1L inhibitors have been several structural data published no pronounced computational efforts have been yet reported. In these studies a first tentative of multi-stage and LB/SB combined approach is reported in order to maximize the use of available data. Using co-crystallized ligand/DOT1L complexes, predictive 3-D QSAR and COMBINE models were built through a python implementation of previously reported methodologies. The models, validated by either modeled or experimental external test sets, proved to have good predictive abilities. The application of these models to an internal library led to the selection of two unreported compounds that were found able to inhibit DOT1L at micromolar level. To the best of our knowledge this is the first report of quantitative LB and SB DOT1L inhibitors models and their application to disclose new potential epigenetic modulators.

A fully featured COMBINE archive of a simulation study on syncytial mitotic cycles in Drosophila embryos

COMBINE archives are standardised containers for data files related to a simulation study in computational biology. This manuscript describes a fully featured archive of a previously published simulation study, including (i) the original publication, (ii) the model, (iii) the analyses, and (iv) metadata describing the files and their origin. With the archived data at hand, it is possible to reproduce the results of the original work. The archive can be used for both, educational and research purposes. Anyone may reuse, extend and update the archive to make it a valuable resource for the scientific community.

Do Alcohol Relapse Episodes During Treatment Predict Long-Term Outcomes? Investigating the Validity of Existing Definitions of Alcohol Use Disorder Relapse

BACKGROUND

The construct of relapse is used widely in clinical research and practice of alcohol use disorder (AUD) treatment. The purpose of this study was to test the predictive validity of commonly appearing definitions of AUD relapse in the empirical literature.

METHODS

Secondary analyses of data from Project MATCH and COMBINE were conducted using 7 definitions of "relapse" based on drinking quantity within a single drinking episode: any drinking; at least 4/5 drinks for women/men; at least 4.3/7.1 drinks for women/men; at least 6/7 drinks for women/men; at least 6 drinks; at least 7/9 drinks for women/men; and at least 8/10 drinks for women/men. Relapse was used to predict alcohol consumption, related consequences, and nonconsumption outcomes (quality of life, psychosocial functioning) at the end of treatment and up to 1 year posttreatment.

RESULTS

Regression analyses indicated within-treatment relapse definitions significantly predicted end-of-treatment alcohol consumption and alcohol-related consequences. Heavy drinking definitions were generally more predictive than the any drinking definition, but no single heavy drinking definition was consistently a better predictor of outcomes. Relapse definitions were less predictive of longer-term alcohol-related outcomes and both shorter- and longer-term nonconsumption outcomes, including health and psychosocial functioning.

CONCLUSIONS

One particular definition of relapse did not consistently stand out as the best predictor. Advances in AUD research may require reconceptualization of relapse as a multifaceted dynamic process and may consider a wider range of relevant behaviors (e.g., health and psychosocial functioning) when examining the change process in individuals with AUD.

Pretreatment alcohol drinking goals are associated with treatment outcomes

BACKGROUND

A large subset of patients who enter treatment for alcohol dependence report nonabstinent drinking goals (e.g., reduction in drinking) rather than abstinence, and this pretreatment goal choice may be associated with drinking outcomes and alcohol-related problems.

METHODS

An analysis of the 16-week Combined Pharmacotherapies and Behavioral Interventions (COMBINE) study was conducted to determine the association between self-reported pretreatment drinking goal and drinking outcomes and alcohol-related problems. Participants who reported a nonabstinent drinking goal (n = 340) were matched with participants who reported an abstinent drinking goal (n = 340) on 3 variables believed to contribute to treatment outcomes: COMBINE experimental group, gender, and number of prebaseline heavy drinking days.

RESULTS

Analyses revealed no interaction between the COMBINE experimental group and drinking goal on outcome measures, so results were collapsed and examined as a function of drinking goal group. Participants who chose an abstinent drinking goal had significantly more weeks with no drinking or no heavy drinking, reported fewer heavy drinking days, reported fewer days with >1 drink, and were more likely to have a ≥50% decrease in drinks per day between baseline and week 16 of the intervention. However, both groups reported reductions over time in percent drinking days, mean drinks per day, number of heavy drinking days, and number of drinking days per week, and participants in both groups experienced significant reductions in alcohol-related problems and improvements in psychosocial functioning.

CONCLUSIONS

Results replicate and expand upon previous studies examining the association between drinking goal and treatment outcome. These data also provide support for the standard inclusion of drinking treatment goal as a stratification variable in study interventions or as a covariate in outcome analyses and highlight several areas that warrant additional research regarding patients who enter alcohol treatment with a nonabstinent drinking goal.