Will machine learning methods such as performing random forest on compositional data (as we have relative abundances for the genera) require any type of data transformation?
@@Riffomonas i see. what I am thinking of and facing is that i have some entries like a code D alphabit d digits DDD-ddd-DDD this patter could repeat fully or partially e.g. DD-dd-DDD-ddd-dddd DDD-ddd-DDD-DDD-ddd-DDD the delimiter is always there
@@ahmed007Jaber Ah, maybe try using the separate function. If you look back through past episodes I have one in there on how to use it. Basically, it lets you separate a column by a delimeter into new columns
@@Riffomonas HI Pat, thank you for the effort you put in replying and sharing knowledge. your comments clicked with me. did look quickly at the aforementioned video and tried the following in regex to match another pattern str_extract(x, pattern= "(........) | (......)") it worked like a charm
What part of working with regular expressions do you find the most frustrating?
HI Pat, great presentation. I love your tutorials. Many thanks for your generous gift.
I learnt about rename_all() for the first time. Thank you.
Wonderful! So glad you learned something new. I mainly use tolower, but you can use any function with rename_all
Great example of real use for manipulating data. Thanks!
Thanks Victor! 🤓
Clear and purpose-driven!
I am loving it - just found about the channel!
So happy, it worked! And I had a completely different raw data - from dada2!
Thanks Pat. Really great tutorial.
Hi Pat, great tutorial! Thanks a lot for this :)
Nice job.
Will machine learning methods such as performing random forest on compositional data (as we have relative abundances for the genera) require any type of data transformation?
We’ll do some transformations but not for compositionality. Depending on your application that may be something to experiment with
@@Riffomonas Thanks, will wait for the tutorials. Your videos are really helpful :)
Hi Pat, thank you for this. Is there a way to specify in the regular expression
if you don't find this regex, look for 2nd form, 3rd form etc???
Tha is for watching! Sorry. Not that I’m aware of other than setting up multiple separate searches
@@Riffomonas i see. what I am thinking of and facing is that i have some entries like a code
D alphabit
d digits
DDD-ddd-DDD
this patter could repeat fully or partially e.g.
DD-dd-DDD-ddd-dddd
DDD-ddd-DDD-DDD-ddd-DDD
the delimiter is always there
@@ahmed007Jaber Ah, maybe try using the separate function. If you look back through past episodes I have one in there on how to use it. Basically, it lets you separate a column by a delimeter into new columns
@@Riffomonas thank you; will check it
@@Riffomonas HI Pat, thank you for the effort you put in replying and sharing knowledge. your comments clicked with me.
did look quickly at the aforementioned video and tried the following in regex to match another pattern
str_extract(x, pattern= "(........) | (......)")
it worked like a charm
Could you introduce how to clean up and manipulate taxnomy data output from qiime2?
Hah! I likely have a conflict of interest, but I’d strongly encourage you to work with your data in mothur instead
You look and sound like a smart version of Seth Rogen haha
The presenter appears to know R; however he doesn't seem to know that parentheSIS is the singular for parentheSES. 😮
Sorry for my midwestern drawl that tends to mangle some words. At least I know that “data” is plural though. #amirite?
It's all good. No problem.