User login
Photo by Darren Baker
Scientists say they have developed a computer program that can predict whether or not a given pharmaceutical agent will produce certain side effects.
The software takes an “ensemble approach” to assessing the chemical structure of a drug molecule and can determine whether key substructures are present in the molecule that are known to give rise to side effects in other drugs.
Md Jamiul Jahid and Jianhua Ruan, PhD, both of the University of Texas at San Antonio, developed the computer program and described it in the International Journal of Computational Biology and Drug Design.
The pair tested the software’s ability to predict 1385 side effects associated with 888 marketed drugs and found that the program outperformed earlier software.
The team also used their new software to test 2883 uncharacterized compounds in the DrugBank database. The program proved capable of predicting a wide variety of side effects, including some effects that were missed by other screening methods.
The scientists believe their software could be used to alert regulatory authorities and healthcare workers as to what side effects might occur when a new drug enters late-stage clinical trials and is ultimately brought to market.
But the program may have an additional benefit as well. By identifying substructures that are associated with particular side effects, the software could be used to help medicinal chemists understand the underlying mechanism by which a side effect arises.
The chemists could then eliminate the offending substructures from drug molecules in the future, thereby reducing the number of drugs that go through the research and development pipeline and then fail in clinical trials due to severe side effects. 
Photo by Darren Baker
Scientists say they have developed a computer program that can predict whether or not a given pharmaceutical agent will produce certain side effects.
The software takes an “ensemble approach” to assessing the chemical structure of a drug molecule and can determine whether key substructures are present in the molecule that are known to give rise to side effects in other drugs.
Md Jamiul Jahid and Jianhua Ruan, PhD, both of the University of Texas at San Antonio, developed the computer program and described it in the International Journal of Computational Biology and Drug Design.
The pair tested the software’s ability to predict 1385 side effects associated with 888 marketed drugs and found that the program outperformed earlier software.
The team also used their new software to test 2883 uncharacterized compounds in the DrugBank database. The program proved capable of predicting a wide variety of side effects, including some effects that were missed by other screening methods.
The scientists believe their software could be used to alert regulatory authorities and healthcare workers as to what side effects might occur when a new drug enters late-stage clinical trials and is ultimately brought to market.
But the program may have an additional benefit as well. By identifying substructures that are associated with particular side effects, the software could be used to help medicinal chemists understand the underlying mechanism by which a side effect arises.
The chemists could then eliminate the offending substructures from drug molecules in the future, thereby reducing the number of drugs that go through the research and development pipeline and then fail in clinical trials due to severe side effects.
Photo by Darren Baker
Scientists say they have developed a computer program that can predict whether or not a given pharmaceutical agent will produce certain side effects.
The software takes an “ensemble approach” to assessing the chemical structure of a drug molecule and can determine whether key substructures are present in the molecule that are known to give rise to side effects in other drugs.
Md Jamiul Jahid and Jianhua Ruan, PhD, both of the University of Texas at San Antonio, developed the computer program and described it in the International Journal of Computational Biology and Drug Design.
The pair tested the software’s ability to predict 1385 side effects associated with 888 marketed drugs and found that the program outperformed earlier software.
The team also used their new software to test 2883 uncharacterized compounds in the DrugBank database. The program proved capable of predicting a wide variety of side effects, including some effects that were missed by other screening methods.
The scientists believe their software could be used to alert regulatory authorities and healthcare workers as to what side effects might occur when a new drug enters late-stage clinical trials and is ultimately brought to market.
But the program may have an additional benefit as well. By identifying substructures that are associated with particular side effects, the software could be used to help medicinal chemists understand the underlying mechanism by which a side effect arises.
The chemists could then eliminate the offending substructures from drug molecules in the future, thereby reducing the number of drugs that go through the research and development pipeline and then fail in clinical trials due to severe side effects.