How to upload my own database on CMAP

Question

[Beowulf]

I am a freshman of biology and just a year2 student of clinical medicine. Nowadays, I am trying to run some data by CMAP(connectivity map), but it doesn't work. The probe sets I want to run can't be found in the database and seems like below:

   down tags: probe '205702_at' invalid, not found in database
          up tags: probe '212733_at' invalid, not found in database
        down tags: probe '216937_s_at' invalid, not found in database

I have no idea about it, and actually I have many other questions about CMAP. I will be grateful if anyone can answer my question. Please help me.

[Bioinformatist Newbie]

Probably the gene signature query you have is from some other organism than Human. CMap just contains the human cancerous cell lines so to query CMap with your signature you have to use Human gene identifiers.

[Beowulf]

The probes I want are HG-U133A platform and I use human liver cancer cell line HepG2. I have read the instruction and run the preload signature, I know the principle of the software. But when I tried to run my probes, it said the probes can not be found in the database, so I guess I should upload something to make my probes recognizable. Actually, I haven't do any IT analyse before but Gsea, I am not really understand what should I do now. I even don't know should I tell the software the specific value of different genes in my management, if so, how can I do it? Maybe my questions are very silly, but I do have no idea. As for the instructions, I have read it many times, but English isn't my mother language, also, some IT terms make me confused. I guess a demonstration will be better and will help me to understand it more. I will be appreciate to you if you can tell me more details.

Answer 1

[Bioinformatist Newbie]

Probably the gene signature query you have is from some other organism than Human. CMap just contains the human cancerous cell lines so to query CMap with your signature you have to use Human gene identifiers. And you question title seems wrong. You cannot upload your own databse on CMap, you can upload a specific Gene signature Query. 2 different files with .grp extension. 1 containing the identifiers for up regulated genes and 1 containing the identifiers for down regulated genes.

Answer 2

[Zhilong Jia]

The probe should be the memory of probes in HG-U133A platform.

A list of probe sets from the feature set representing either the 'up' genes in a signature (the 'up tag list') or the 'down' genes in a signature (the 'down tag list').

The universe of probe sets recognized by cmap, defined as the 22,283 probe sets on the HG-U133A array

ref. http://www.connectivitymap.org/cmap/help_topics_frames.jsp?topic=tag%20list

[Beowulf]

The probes I want are HG-U133A platform and I use human liver cancer cell line HepG2. I have read the instruction and run the preload signature, I know the principle of the software. But when I tried to run my probes, it said the probes can not be found in the database, so I guess I should upload something to make my probes recognizable. Actually, I haven't do any IT analyse before but Gsea, I am not really understand what should I do now. I even don't know should I tell the software the specific value of different genes in my management, if so, how can I do it? Maybe my questions are very silly, but I do have no idea. As for the instructions, I have read it many times, but English isn't my mother language, also, some IT terms make me confused. I guess a demonstration will be better and will help me to understand it more. I will be appreciate to you if you can tell me more details.

[Zhilong Jia]

When I used the three probes to query cmap, it works for me. So check your two grp files, like format etc. And try the example data set.

[Beowulf]

When you run your data, does the grp file you used contain the specific value or you just use the probe set?

[Zhilong Jia]

just probe. I suggest you run the example firstly.

[Beowulf]

I have run it and I succeed. But my data can still not work.

[Zhilong Jia]

please paste your two files in this post.

[Beowulf]

down tag

up tag

[Zhilong Jia]

No more than 1000 probes, containing up and down-regulated genes, are supported.

A list of probe sets from the feature set representing either the 'up' genes in a signature (the 'up tag list') or the 'down' genes in a signature (the 'down tag list'). One up tag list and one down tag list together represent a signature and both are required to execute a query using the quick query page or to upload a signature using the load signature page. Note that the total number of probe sets in the up tag list and down tag lists may not together exceed 1000. Text files with the .grp extension are used to contain tag lists. A .grp file has one probe set on each line.

Ref: http://www.connectivitymap.org/cmap/help_topics_frames.jsp?topic=tag%20list

[Beowulf]

OK I will select the probes and try it again. Thanks for your help. May I ask other questions if there are any other question? Thank you again.

[Beowulf]

I have changed the number of probes and they are less than 1000, but they can still not be recognized. I guess HepG2 cell line is not contained in the database. Do you know any method to deal with this problem? Thank you.

[Beowulf]

Now I succeed! Thank you, really thank you! I don't know what to say now but thank you.

[Beowulf]

But there is still a question, the cell line I use is HepG2, while the result shows many cell lines but HepG2. Do you think that means the result will be useless in my analysis? Or do you think the result is still creditable?

[Zhilong Jia]

There are 3/4 cell lines in CMAP in total. Generally, the cell line may be not considered as there is a robust signature from different cell lines in most cases. I suggest your read the original cmap paper and others' using cmap. Ref: http://www.connectivitymap.org/cmap/publications.jsp

[Bioinformatist Newbie]

Zhilong, Do you have any idea how we can analyse the entire CMap dataset? For furthe details, kindly see this post: Differential Expression Analyses of Entire Connectivity Map (microarray) dataset

[Zhilong Jia]

Actually, I did mark your question for waiting for others' answers as mine is incomplete.

[Beowulf]

OK, thanks for your help. I have discussed it with my boss and we choose another plan, so I don't need to do it any more. Thank you, again.

[Beowulf]

When you run your data, does the grp file you used contain the specific value or only have the probe sets?