Quantitative structure–activity relationship (QSAR) models experts

• Andrea Mauri

  a new tutorial "Variable selection methods: an introduction" as been uploaded on moleculardescriptors.eu website

  http://moleculardescriptors.eu/tutorials/tutorials.htm
• Kunal Roy

  The Extended Topochemical Atom (ETA) indices developed by the Drug Theoretics Lab, Jadavpur University ( https://sites.google.com/site/kunalroyindia/ ) are now available for computation inThe Extended Topochemical Atom (ETA) indices developed by the Drug Theoretics Lab, Jadavpur University ( https://sites.google.com/site/kunalroyindia/ ) are now available for computation in PaDEL-Descriptor, an open source software of NUS, Singapore ( http://padel.nus.edu.sg/software/padeldescriptor/ ).

  Recent replies ⋅ Show All (3)

  • 
    Subhadip Banerjee replied

    Really of great help sir.

    Apr 22, 2012
• Yogesh Wagh

  VLife Research Publication Contest VLife now opens the entries for VLife Best Research Publication 2011, for outstanding publication in the field of computer aided drug and molecular design.

  http://www.vlifesciences.com/support/BestPublication.php
• Mukesh Yadav

  what is the impact of biological activity dispersion on statistical fitness of QSAR models? is there any method available to determine it?

  it is known that biological activities in QSAR are termed as dependent variable in regression. numerical variation in activities definitely alters the statistical fitness of QSAR of models. how canit is known that biological activities in QSAR are termed as dependent variable in regression. numerical variation in activities definitely alters the statistical fitness of QSAR of models. how can we calculate and regulate activity magnitudes to obtain statistically fit model?

  Recent replies ⋅ Show All (5)

  • 
    Oleg Devinyak replied

    Generally saying, you can use any kind of experimental data as a dependent variable. Using logarithmic unit of molar concentration gives you only slightly (and I think even negligible) higher chanceGenerally saying, you can use any kind of experimental data as a dependent variable. Using logarithmic unit of molar concentration gives you only slightly (and I think even negligible) higher chance to find linear relationship. And not the activity dispersion itself, but the ratio of activity dispersion/experimental error limits your goodness-of-fit. Generally you can use the next inequality R^2<1-(SE/SD)^2 (SE-standard error of experiment, SD - standard deviation of your activity sample). This inequality says that you cannot obtain a model that is more exact than experiment it is based on. If your R2 is higher then overfitting is obvious. //(I have missed the square, so the post is corrected now)

    Mar 23, 2012
• Yogesh Wagh

  VLife Research Publication Contest VLife now opens the entries for VLife Best Research Publication 2011, for outstanding publication in the field of computer aided drug and molecular design.

  http://www.vlifesciences.com/support/BestPublication.php
• Yogesh Wagh

  How useful are the modeled protein structures?

  http://blog.vlifesciences.com/?p=255

  Recent replies ⋅ Show All (3)

  • 
    Hendy Pryanda replied

    thanks for your recomend sir! I want to try a prion modeling! do you ever heard this protein? can I do that with VLife?

    Mar 22, 2012
• Kunal Roy

  Special Issue of Current Drug Safety on “In silico modeling for prediction of drug-induced adverse reactions and environmental hazards using QSAR tools” The field of in silico predictiveSpecial Issue of Current Drug Safety on “In silico modeling for prediction of drug-induced adverse reactions and environmental hazards using QSAR tools” The field of in silico predictive toxicology is receiving increased attention because studies have indicated that adverse drug reactions (ADRs) are among the causes of costly late-stage failures in drug development. Pharmaceuticals do not always reveal all side-effects or adverse reactions during the pre-marketing investigation, preclinical and clinical trials of a new medicinal product. The existence of ADRs becomes apparent once the compound has been extensively prescribed and a population base of considerable size has been exposed to the therapeutic agent. This may be due to failure to mimic the factors or co-factors of real life during clinical trial. Considerable factions of potential drug candidates fail because of undesirable toxicity properties. Drug-induced adverse reactions are one of the leading causes of morbidity and mortality, and potential drug-induced toxicity parameters should be identified at the earlier stage of the drug development process as in this case less time and expense will be invested in an ultimate failure. Another important emerging issue is environmental toxicity of pharmaceuticals. Pharmaceuticals being extensively and progressively used in human and veterinary medicine are emerging as significant environmental contaminants. Pharmaceuticals are designed to have a specific mode of action and many of them are persistent in the body. These features among others make pharmaceuticals to be evaluated for potential effects on different representative organisms of the ecosystem. The major routes of entry of pharmaceuticals into the environment are from their use by individuals either dispersed throughout the community or concentrated in medical centres of hospitals, and the disposal of unwanted or expired drugs by users. Low levels of pharmaceuticals have been detected in many countries in sewage treatment plant effluents, surface waters, groundwater and drinking waters. Although pharmaceuticals receive extensive pharmacological and clinical testing, information on the ecotoxicity of these biologically active compounds is in general very limited. Quantitative structure-activity relationships (QSARs) are increasingly being used for the prediction of toxicity properties of drug leads issue as an alternative to non-animal method. Though a number of in vitro and high throughput methods have been developed to provide experimental evaluation of such properties, earliest possible identification of a liability is desired, including even prior to synthesis. Thus, in silico approaches may be used for hypothetical compounds to guide synthetic efforts. Moreover, in silico calculations may be done quickly and a large number of compounds may be screened within very short period. In silico methods may be used to prioritize compounds for synthesis and experimental assay. Considering the importance of in silico modeling for predicting various toxicity endpoints, Bentham Science Publishers have decided to publish a special issue of Current Drug Safety (ISSN: 1574-8863; http://benthamscience.com/cds/) on “In silico modeling for prediction of drug-induced adverse reactions and environmental hazards using QSAR tools”. The planned issue of Current Drug Safety will focus on development of in silico predictive models for various endpoints of drug-induced toxicities including ecotoxicity of pharmaceuticals. Both review and original research papers on QSAR modeling of drug-induced toxicity and ecotoxicity will be considered. If you want to contribute an article (Research Paper/Review Article) to this special issue, kindly send us the tentative title of the review/ research article and a brief (about 200 words) abstract by March 31, 2012. You may go through the Instructions for Authors at http://benthamscience.com/cds/MSandI.htm . We would request the contributors to submit the full papers by June 30, 2012. If you require more time for submission, kindly let us know. All submissions will be peer reviewed by anonymous referees and editors. If you have any further questions, kindly do not hesitate to contact us. Dr. Achintya Saha, Department of Chemical Technology, University of Calcutta, Kolkata 700009, India; achintya_saha@yahoo.com Dr. Kunal Roy, Department of Pharmaceutical Technology, Jadavpur University, Kolkata 700032, India; kunalroy_in@yahoo.com [Guest Editors]
• Nirzari Gupta

  reliable method for outlier detection and elimination in 3D QSAR CoMFA?

  please anyone suggest the outlier detection and elimination method in 3D QSAR CoMFA using Sybyl.

  Recent replies ⋅ Show All (4)

  • 
    Nirzari Gupta replied

    thank you david. Nice guidance.

    Feb 25, 2012
• Kunal Roy

  QSAR/QSPR researchers may find this publication interesting http://pubs.acs.org/doi/abs/10.1021/ci200520g
• Yogesh Wagh

  Multicore support in VLifeMDS- Faster and Efficient molecular modeling

  http://blog.vlifesciences.com/?p=249
• Yogesh Wagh

  VLifeMDS, especially the popular VLife GQSAR has now got another eminent user USM, Malaysia. If you would like to know more about GQSAR please write to me

  Recent replies ⋅ Show All (1)

  • 
    Ramalakshmi Kota replied

    Hello sir.....i would lyk 2 knw abt qsar in general ....can u suggest some books 4m where can v understand in simple language

    Jan 9, 2012
• Kunal Roy

  Validation of QSAR models

  Use of rm2 metrics in validation of QSAR models

  https://sites.google.com/site/rm2forqsarvalidation/

  Recent replies ⋅ Show All (1)

  • 
    Bangaru Bhaskarara Rao replied

    Thank you very much sir, for giving details.

    Jan 3, 2012
• Emilio Esposito

  Thoughts and comments on Open3DQSAR and Open3DALIGN

  Hi Does anyone have experience using Open3DQSAR (http://open3dqsar.sourceforge.net/) and Open3DALIGN (http://open3dalign.sourceforge.net/)? Specifically, how do the results these applicationsHi Does anyone have experience using Open3DQSAR (http://open3dqsar.sourceforge.net/) and Open3DALIGN (http://open3dalign.sourceforge.net/)? Specifically, how do the results these applications generate compare to CoMFA and CoMSIA? Thank you

  Recent replies ⋅ Show All (1)

  • 
    Paolo Tosco replied

    Dear Emilio, at the DOI http://dx.doi.org/10.1021/ci200411s you may find a thorough validation of the Open3DQSAR/Open3DALIGN suite. Specifically, results obtained with Open3DQSAR on the sameDear Emilio, at the DOI http://dx.doi.org/10.1021/ci200411s you may find a thorough validation of the Open3DQSAR/Open3DALIGN suite. Specifically, results obtained with Open3DQSAR on the same alignments used for CoMFA are definitely comparable. Best regards, Paolo

    Dec 12, 2011
• Nini Guaia

  Is it possible to use molar absorptivity as a QSAR descriptor?

  Can molar absorptivity at a determinate UV wavelenght be used in a QSAR equation? Why? To which parameters is not orthogonal? Thanks!

  Recent replies ⋅ Show All (3)

  • 
    Saeed Yousefinejad replied

    The above comment could not satisfy me . I think we can use all structural properties of molecules as a independent descriptor, but we should care that it should not have a serious correlation withThe above comment could not satisfy me . I think we can use all structural properties of molecules as a independent descriptor, but we should care that it should not have a serious correlation with the other applied descriptors.and we know that molar absorptivity is depended on the molecular structure. but about molar absorptivity we may use it only for the molecules with very very similar structures and not a diverse data set. because different structures might be active in the same wavelength and take the problem more complex.we could not say we can not use an experimental parameter like molar absorptivity only because its related to the analysis. its just my idea...

    Nov 27, 2011
• Yogesh Wagh

  Webinar recording - GQSAR for GPCR studies

  http://vimeo.com/user9121494/gpcr