search

StanfordBioinformatics / pulsar_lims

A LIMS for ENCODE submitting labs.
3 stars 1 forks source link

Encode data submission: missing antibody for cs-300 #406

Open twang15 opened 3 years ago

twang15 commented 3 years ago

https://www.encodeproject.org/experiments/ENCSR312HLG/

twang15 commented 3 years ago

Hi Annika,

Could you help me register this antibody https://pulsar-encode.herokuapp.com/antibodies/24 on the portal?

Thanks, Tao

twang15 commented 3 years ago

It’s on the portal already: ENCAB909IOP (I probably forgot to add it to Pulsar)

But it was rules non-compliant. This is the IP that Jie is working on.

twang15 commented 3 years ago

Hmm, could you explain a bit about what does “it was rules non-compliant” mean? Shall I wait for a new antibody or sth else?

Thanks, Tao

twang15 commented 3 years ago

From Annika:

Yes. We did some experiments (Histone ChIP) and for those we need to validate the antibody with a western and a dot plot. I uploaded those validations to the portal and while they are not perfect they are certainly good enough in our opinion to get through the review process (by the antibody characterization working group). However, they confirmed the compliance for 1 out of the 3. But they want us to repeat two of them. You can see they status in this spreadsheet:

https://docs.google.com/spreadsheets/d/1ZjBGXzIjtrN-MiczmehK-MoYbybKvEan4_TJ9eYfinw/edit#gid=0

So I asked Jie to repeat the two that are not compliant. You can ignore this issue for now and I can just upload the IP when Jie is done. Unfortunately it keeps quite some data sets from being released.

twang15 commented 3 years ago

Hi Annika,

Here are several questions I have: What is antibody characterization? Relation to IP (biosample characterization) and PCR (GM characterization)? What to do about those experiment (in purple color) in the list: https://docs.google.com/spreadsheets/d/13mhYXu0ufgGzlc_vCuQkBbwVno5eFzYgfpNUZd9HU04/edit#gid=469741889

Thanks, Tao

twang15 commented 3 years ago

Hi Annika,

Could you explain a bit the following questions for me?

Thanks, Tao

twang15 commented 3 years ago

Yes, I will. Sorry for the late reply. I didn’t forget about it, just didn’t find time yet.

  1. When we use an antibody to pull a certain protein, for example a GFP antibody to pull down TFs that are tagged by GFP, we need to proof that our antibody is actually capable of pulling down the correct protein and not some random protein (however there’s usually a bit of background). Usually we only use the GFP antibody for our ChIP experiments and we did this validation once and it’s good for all experiments because we use the same antibody for all of them. However, the cases you see on the portal now are not from a GFP antibody but a histone antibody. This one binds to certain epigenetic modifications (histone modifications) on histone (a protein that binds to DNA, or more precisely the DNA is wrapped around histones) and pulls down the part of the DNA that is modified (—> example H3K4me3: this one binds to histone 3 (H3), at the K (=lysin) amino acid on position 4 of the protein (K4) if it has three methylated residues (me3). This indicates certain gene regulatory activity: there are activatory (like H3K4me2) and repressive histone modifications H3K27me3). Since we didn’t use those antibodies before we have to show that they are specific (pull down the right protein, right size, with a western blot).

  2. Relation to IP and PCR: none

  3. We cannot do anything about the purple experiments in the list right now because we have to wait for Jie’s validation of the histone antibody that she has to repeat. Only if the antibody passes, the DCC will release the experiments and the audit will go away.