Hi, I am a beginner in bioinformatics and I am using homology modelling to predict the tertiary structure of a protein sequence. I am using SWISS-MODEL to construct the tertiary structure. However, I have also seen in many posts online that MODELLER is a tool that is used a lot to predict the tertiary structure of a protein. I would like to know that whether these two tools are as equally as good as each other, or are there certain advantages in using one of these tools over the other? Further, with regards to validating the structure created through homology modelling, I have read that the tools that you will need to use to validate the structure will differ depending on whether SWISS-MODEL or MODELLER is used. Is this correct? And when it comes to validating your structure, are there certain advantages in using one tool over the other to create the tertiary structure (e.g. SWISS-MODEL vs MODELLER)? I would really appreciate any insights from someone who has experience using these tools. Thanks.
Look up the results of the CASP competitions.
They rank simulation software. Different tools may be a bit better for different proteins.
Both seem to me to be excellent tools, though you can always try i-tasser which has been in the 1st place in CASP for years.
I-TASSER is my go-to choice (mainly for that reason)
They are both excellent modelling programs. I wouldn't worry about any difference in quality. At the end of the day every model will require refinement after using evaluation tools. No modelling software will get it right first time.
Swiss model, because of the online user interface, is far more user friendly than modeller and you may find modelling bound ligands easier also. However it only returns a single result and is of course dependent on internet access. Also, a nice thing is that you get an image of the result immeadiately without further work.
Modeller is a command line tool so requires more involvmenent. The installation is simple, but there are a few steps involved before you can get it to work i.e. setting up the .ali and .py files. Make notes while you do this! The result provides 5 different models and you can evaluate each of them or simply choose the one with the lowest free energy (which is provided in the log file). The results don't provide an image, which is fine because you will want to visualise and manipulate the model in something like Pymol anyway.
Personally I only use Modeller.