Home Submit Algorithms Help Team Contact

Inhibitor Prediction

  KiDoQ (Mtb target)

  GDoQ (Mtb target)

  ABMpred (Mtb target)

  eBooster (Mtb target)

  MDRIpred (Mtb cell)

  CancerIN (Cancer)

  ntEGFR (Cancer EGFR)

  EGFRpred (Cancer EGFR)

  DiPCell (Pancreatic Cancer)

  DMKPred (Human Kinases)

  TLR4HI (Human TLR4)

  HIVFin (HIV)

Antigenic Properties

ADMET Properties

  MetaPred (Cytochrome P450)

  ToxiPred (Aqueous toxicity)

  DrugMint (Drug-like)

  QED (Oral drug-like)

  Format Conversion



Toll like Receptors (TLRs) which are characterized by the presence of cytoplasmic TLR domain is a very important patteren recognistion receptors. At present ~10 members of TLR family have been identified in Human that plays important role in Cell Signaling process. Lipopolysaccharides (LPS) is a pathogenic molecules which are recognized by innate immunity system (first line of defense). Action of Lipopolysaccharides is associated with TLR4 Signaling. Thus unbalanced/improper TLR4 signaling is associated with several pathological diseases.

The current knowledge of the structure and function of TLR4 signaling has opened the possibility to develop new inhibitors against TLR4. Bioassays and other experimental protocol to identify TLR4 inhibitors were already known but require both huge money and labor therefore a computational approach is needed to assist in the screening of chemical compounds on which further validation can be done. In the present study, a systematic attempt has been made to address these challenges. We have taken a confirmatory bioassay compounds dataset which have both active and inactive compounds. An efficient QSAR models was developed with commonly used molecular descriptors, calculated using commercial and freely available software packages. We applied all possible machine learning techniques and based on the best models developed a user friendly web server "TLR4HI" and hence providing an open source platform for scientific community.

The QSAR model developed shows a correlation cofficient(R)/R2 0.74/0.55 with RMSE 22.30 using 5 fold cross validation techniques with freely available PowerMV software.