golgi | library built on top of Gorgonia | Machine Learning library

As commented, your earlier search returns an empty data frame since the searched term is part of the larger strings in the character column, GO_NAME. Therefore, instead of == or%in% operators which expects whole word matches, consider grep to search string patterns within larger string:

If you use ggplot and bring your data into a tidy form (i.e. long format), you can use easily use ggplot to plot several boxplots at once, as suggested in the comments.

I'm not sure if I completely understand what you're after, but if you want one boxplot per compartiment, combining the gene expression of all genes present in said compartiment, this is a possible solution (I didn't use all your data as it is quite an effort to get it into R in the way you present it):

Here is a script to clean the column. Note you may want to add more words to the stopword set to meet your requirements.

first a little advice, separate php, javascript and html. Together it is difficult to read and maintain. Otherwise you create a function at each pass of the loop. It doesn't bother me that nothing works like that. Try it like this:

Send the id of your image as a parameter to the function. A single function, which will be after the loop.

You could output the function to create a list-of-lists

AWK:

If you don't mind, you can use the rvest package:

This may be one way to go. Since your expected result is a character vector, I cannot visualize the final output. Yet, you said you want to check how many columns each protein appears in in the data. I hope the outcome I have is what you are after.

First, I converted all columns to character. Then, I converted the data to long format one using gather(). For each subcellular structure group (i.e., subcellular), I added row indices (e.g., 1 means the 1st row in your original data), and trim white space. Then, remove any rows with NA in protein. Remove any rows with "" and " ". Now tidying up is done. For each row (i.e., row.index), remove duplicated protein types. Ungroup the data, and finally count how many columns each protein appears (i.e., sucellular structure). Basically, you want to count how many times each protein appear in the data set by this time.

With your sample data, I got the following result. But I am not sure if this is what you want. (I am off to bed now. So I cannot help you for some hours. If anybody can jump in, please do so.)

We can use str_count. Loop over the columns except the first one (lapply(df[-1], ..), get the count of ; add 1 to it, check for cases where there is empty string and replace those elements with NA