Open InvincibleZZH opened 1 year ago
do I need your dataset to predict the important site in my own protein?
No description provided.
and thank you very much for your instructions! it helped my research a lot!
Not at all. It's just an example of how I formatted it.
Not at all. It's just an example of how I formatted it.
so where can i download your dataset?
I'm super curious what you're working on. I'm happy to help in any way I can!
iSero RF data (Round 0) is probably the tab you want to start with.
thank you very much! there are also two issues troubling me, I still cant edit a right xxxx.res in step two backrub 😭
I'm super curious what you're working on. I'm happy to help in any way I can!
1129804456@qq.com this is my email, can we talk about my research through this privately,sir?
Unfortunately my labmate is the one who did all the Rosetta modeling, so I know a lot less about that. I'm not sure if he's on Github.
Unfortunately my labmate is the one who did all the Rosetta modeling, so I know a lot less about that. I'm not sure if he's on Github.
thats OK!I will keep trying to get a right file,its quiet interesting and challenging!
I used the random forest model through R. There is a package called "rattle" (if you install it, there are quite a few dependencies, and some might be out of date, so you might have to manually install updated dependencies). Rattle is a GUI, so it's pretty easy to use. On the first tab you upload your data. My data is available if you want to use it or see the formatting. There is another tab where you can generate models for your data. Random forest is one of the options, and you can set different parameters. The parameters I used are listed in the paper and below, but there's nothing special about them, I just played with different numbers until I got results that made sense and that I could test.
You can also run a random forest directly through R using the package randomForest https://cran.r-project.org/web/packages/randomForest/randomForest.pdf
https://rattle.togaware.com/
randomForest(formula = x5HT ~., data = crs$dataset[, c(crs$input, crs$target)], ntree = 500, mtry = 14, importance = TRUE, replace = FALSE, na.action = randomForest::na.roughfix) Type of random forest: regression Number of trees: 500 No. of variables tried at each split: 14 Missing value imputation is active.