Open Bhersbach opened 1 year ago
Yes, I'm sure we can solve this quickly. Have you defined the dataset hierarchy as explained here or not: https://cellbrowser.readthedocs.io/en/develop/collections.html
Does your instance have a publicly accessible URL so I can have a quick look?
On Mon, Sep 12, 2022 at 12:18 PM Bhersbach @.***> wrote:
We are setting up Cell Browser in our lab and have successfully prepared two Cell Browser folders as a test. However, when we use cbBuild to host the data, we can access the individual datasets only by using the dataset name, i.e., server address:/?ds=dataset_name. The index/title page, does not load and, instead, we get an empty Cell Browser page. When attempting to open a dataset through file > open dataset, no options are given either. Is there something we are forgetting to specify?
System specs: OS: Linux (Ubuntu 18.04) Cell Browser v1.2.1
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TNNS5KWJJ6I3LVNMETV537OPANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you are subscribed to this thread.Message ID: @.***>
I hadn't done that and just gave it a try, it doesn't seem to have solved the issue (or I did it incorrectly). Unfortunately, our instance is currently not public, so I will try to briefly summarize the situation:
export CBDATAROOT='/home/cellbrowser/User1/'
My ~/.cellbrowser.conf looks like this:
gaTag = "G-5CVR5ZC288" # test, GA3 and GA4 combined htmlDir = "/usr/local/apache/htdocs-cells" dataRoot = "/hive/data/inside/cells/datasets/" outDirs = {"alpha" : "/usr/local/apache/htdocs-cells", "beta" : "/usr/local/apache/htdocs-cells-beta/" } reqTags = ["body_parts", "unit"] # you won't need this email = @.***"
matrixFormat = "mtx" # default matrix format
On Mon, Sep 12, 2022 at 3:42 PM Bhersbach @.***> wrote:
I hadn't done that and just gave it a try, it doesn't seem to have solved the issue (or I did it incorrectly). Unfortunately, our instance is currently not public, so I will try to briefly summarize the situation:
- We have one folder (User1) that contains two datasets (Dataset1 & Dataset2)
- The ~/.cellbrowser.conf file does not seem to be present the home folder so I ran export CBDATAROOT='/home/cellbrowser/User1/'
- I copied cellbrowser.conf and desc.conf from Dataset1 to the User1 folder and edited as described in the link
- I run cbBuild -r /home/cellbrowser/html -p 8999 within the User1 folder, this runs but does not start a CellBrowser server
- I tried running the same command within the Dataset1 folder, which runs and creates a CellBrowser server but no index/title page
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1243763380, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TJ5FSF55MF63RFMO33V54XLZANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Thanks! My home directory doesn't have a .cellbrowser.conf file though. Is it located somewhere else when using a conda environment by any chance?
How about just creating a config file?
echo 'dataRoot = "/home/cellbrowser/User1/"' > ~/.cellbrowser.conf
I think I'll add this command to the documentation...
On Mon, Sep 12, 2022 at 4:00 PM Bhersbach @.***> wrote:
Thanks! My home directory doesn't have a .cellbrowser.conf file though. Is it located somewhere else when using a conda environment by any chance?
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1243787468, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TOF55EMZZ34WSSI3RDV54ZRRANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Done, but to no avail. Also the command cbBuild -r -o /home/cellbrowser/html -p 8999 runs but does not yields a message like: Serving /home/cellbrowser/html/ on port 8999 Point your internet browser to http://127.0.0.1:8999 (or the IP address of this server)
This should be enough:
cd /home/cellbrowser/User1 cbBuild -o /home/cellbrowser/html -p 8899
In /home/cellbrowser/html there should be a desc.json file then with your html text. and a dataset.json file.
On Mon, Sep 12, 2022 at 4:48 PM Maximilian Haeussler @.***> wrote:
Yes, please don't use the -r option, it's not meant to be used with the -p option, sorry I didn't mention that (and it's not documented I just saw).
No need for the -r just run cbBuild with the -o and -p options.
Also, if this is a production webserver, don't use the -p built-in server, it's slow and can use a lot of RAM. I'd rather use any other webserver, apache, nginx or anything else you like.
On Mon, Sep 12, 2022 at 4:29 PM Bhersbach @.***> wrote:
Done, but to no avail. Also the command cbBuild -r -o /home/cellbrowser/html -p 8999 runs but does not yields a message like: Serving /home/cellbrowser/html/ on port 8999 Point your internet browser to http://127.0.0.1:8999 (or the IP address of this server)
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1243829271, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TKLQMH7XFDYII3ETB3V5446LANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
I ran the code as you described. The webserver it launches is empty, I cannot access any datasets even by their url. Going to the html folder and opening the index html gives the following error: Could not find a dataset at dataset.json?86f083f58d. Would it make sense to do a video call so I show you?
We don't usually do video calls, also I don't know your time zone or affiliation. We could do that, but as a last resort.
Yes it cannot find the dataset.json file.
Did you check ~/.cellbrowser.conf, what is in the file? Can you copy it here?
Did you check this: "In /home/cellbrowser/html there should be a desc.json file then with your html text. and a dataset.json file."
On Tue, Sep 13, 2022 at 9:16 AM Bhersbach @.***> wrote:
I ran the code as you described. The webserver it launches is empty, I cannot access any datasets even by their url. Going to the html folder and opening the index html gives the following error: Could not find a dataset at dataset.json?86f083f58d. Would it make sense to do a video call so I show you?
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1244998500, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TPPUWZUTWVDIDWWBRLV6AS4RANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
So our ~/.cellbrowser.conf file contains a single line: dataRoot = "/home/cellbrowser/OutputData/Bob/" Within the html folder there are indeed the desc.json, the inded.html and dataset.json file
Previously you wrote that you used this command:
CBDATAROOT='/home/cellbrowser/User1/
But in the config file you have:
dataRoot = "/home/cellbrowser/OutputData/Bob/"
looks odd to me, are the data files of dataset1 and dataset2 really located under /home/cellbrowser/OutputData/Bob/ ? So you have directories such as /home/cellbrowser/OutputData/Bob/dataset1 and /home/cellbrowser/OutputData/Bob/dataset2?
And you have the top-level cellbrowser.conf and a desc.conf with a general description of your overall webserver in the directory "/home/cellbrowser/OutputData/Bob/" ?
On Tue, Sep 13, 2022 at 11:36 AM Bhersbach @.***> wrote:
So our ~/.cellbrowser.conf file contains a single line: dataRoot = "/home/cellbrowser/OutputData/Bob/" Within the html folder there are indeed the desc.json, the inded.html and dataset.json file
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1245151690, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TO4X2C5RXP5HAUMGJTV6BDIFANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Yes, sorry that was just for privacy reasons. I indeed have the following two folders: /home/cellbrowser/OutputData/Bob/Dataset1 /home/cellbrowser/OutputData/Bob/Dataset2 These contain the objects created with cbImportScanpy
The folder Bob contains the cellbrowser.conf and desc.conf files as per the documentations instructions.
The newly created .cellbrowser.conf file is in /home/cellbrowser/
I can run cbBuild -r within the folder /home/cellbrowser/OutputData/Bob/ and this will generate the html and json files within /home/cellbrowser/html. However, clicking on the html file I get the error previously mentioned. I can also run cbBuild within the dataset folders themselves but that gives a similar outcome.
The data root folder is determined in this code function:
def findRoot(inDir=None): """ return directory dataRoot defined in config file or CBDATAROOT environment variable. """
if 'CBDATAROOT' in os.environ:
dataRoot = os.environ['CBDATAROOT']
else:
dataRoot = getConfig("dataRoot")
if dataRoot is None:
logging.info("dataRoot is not set in ~/.cellbrowser.conf or via
$CBDATAROOT. Dataset hierarchies are not supported.") return None
dataRoot = abspath(expanduser(dataRoot).rstrip("/"))
if inDir is not None and not abspath(inDir).startswith(dataRoot):
logging.info("input directory %s is not located under dataRoot %s.
Deactivating hierarchies." % (inDir, dataRoot)) return None
return dataRoot
Can you look for any of these messages in the cbBuild output and let me know what they are?
On Tue, Sep 13, 2022 at 3:27 PM Bhersbach @.***> wrote:
Yes, sorry that was just for privacy reasons. I indeed have the following two folders: /home/cellbrowser/OutputData/Bob/Dataset1 /home/cellbrowser/OutputData/Bob/Dataset2 These contain the objects created with cbImportScanpy
The folder Bob contains the cellbrowser.conf and desc.conf files as per the documentations instructions.
The newly created .cellbrowser.conf file is in /home/cellbrowser/
I can run cbBuild -r within the folder /home/cellbrowser/OutputData/Bob/ and this will generate the html and json files within /home/cellbrowser/html. However, clicking on the html file I get the error previously mentioned. I can also run cbBuild within the dataset folders themselves but that gives a similar outcome.
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1245413383, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TMQ2SDNM2J7WBYFG63V6B6KZANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
More concretely, can you look for "Deactivating hierarchies" or "Dataset hierarchies are not supported" ?
On Tue, Sep 13, 2022 at 3:30 PM Maximilian Haeussler @.***> wrote:
The data root folder is determined in this code function:
def findRoot(inDir=None): """ return directory dataRoot defined in config file or CBDATAROOT environment variable. """
if 'CBDATAROOT' in os.environ: dataRoot = os.environ['CBDATAROOT'] else: dataRoot = getConfig("dataRoot") if dataRoot is None: logging.info("dataRoot is not set in ~/.cellbrowser.conf or via
$CBDATAROOT. Dataset hierarchies are not supported.") return None
dataRoot = abspath(expanduser(dataRoot).rstrip("/")) if inDir is not None and not abspath(inDir).startswith(dataRoot): logging.info("input directory %s is not located under dataRoot
%s. Deactivating hierarchies." % (inDir, dataRoot)) return None
return dataRoot
Can you look for any of these messages in the cbBuild output and let me know what they are?
On Tue, Sep 13, 2022 at 3:27 PM Bhersbach @.***> wrote:
Yes, sorry that was just for privacy reasons. I indeed have the following two folders: /home/cellbrowser/OutputData/Bob/Dataset1 /home/cellbrowser/OutputData/Bob/Dataset2 These contain the objects created with cbImportScanpy
The folder Bob contains the cellbrowser.conf and desc.conf files as per the documentations instructions.
The newly created .cellbrowser.conf file is in /home/cellbrowser/
I can run cbBuild -r within the folder /home/cellbrowser/OutputData/Bob/ and this will generate the html and json files within /home/cellbrowser/html. However, clicking on the html file I get the error previously mentioned. I can also run cbBuild within the dataset folders themselves but that gives a similar outcome.
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1245413383, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TMQ2SDNM2J7WBYFG63V6B6KZANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Hi. I'm running into same proplem. In my case a computer restart worked and data hierarchy was created, at least inside the html folder I could see 'inDir' and the path to my root data directory in html/dataset.json file. However, when I open the browser by typing python -m http.server 8080
inside the html directory, the UI shows nothing. Under the developer tools, I see the following error message. Even opening the link http://cells.ucsc.edu/ does not show anything anymore.
Ohhh!!! Darn. I can fix this. Yes, we have new "disease filters". Apparently somewhere it assumes that "disease=["cancer"]" or something like this is defined in the cellbrowser.conf file.
Are you using a github clone? If I fix this now, could you use the updated source code form Github or do you need a pip release?
On Thu, Sep 15, 2022 at 8:41 PM Tabassum Kakar @.***> wrote:
Hi. I'm running into same proplem. In my case a computer restart worked and data hierarchy was created, at least inside the html folder I could see 'inDir' and the path to my root data directory in html/dataset.json file. However, when I open the browser by typing python -m http.server 8080 inside the html directory, the UI shows nothing. Under the developer tools, I see the following error message.
[image: Screen Shot 2022-09-15 at 2 39 40 PM] https://user-images.githubusercontent.com/14091615/190483910-f7889565-013e-4687-900f-ed7e0e73746b.png
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1248467511, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TMWHN6S5YD5BOUXAHLV6NUTXANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Pip release please. Thank you.
Get Outlook for iOShttps://aka.ms/o0ukef
From: Maximilian Haeussler @.> Sent: Thursday, September 15, 2022 4:53:08 PM To: maximilianh/cellBrowser @.> Cc: Kakar, Tabassum @.>; Comment @.> Subject: [EXT] Re: [maximilianh/cellBrowser] Question/Issue: Index not showing (Issue #251)
Ohhh!!! Darn. I can fix this. Yes, we have new "disease filters". Apparently somewhere it assumes that "disease=["cancer"]" or something like this is defined in the cellbrowser.conf file.
Are you using a github clone? If I fix this now, could you use the updated source code form Github or do you need a pip release?
On Thu, Sep 15, 2022 at 8:41 PM Tabassum Kakar @.***> wrote:
Hi. I'm running into same proplem. In my case a computer restart worked and data hierarchy was created, at least inside the html folder I could see 'inDir' and the path to my root data directory in html/dataset.json file. However, when I open the browser by typing python -m http.server 8080 inside the html directory, the UI shows nothing. Under the developer tools, I see the following error message.
[image: Screen Shot 2022-09-15 at 2 39 40 PM] https://user-images.githubusercontent.com/14091615/190483910-f7889565-013e-4687-900f-ed7e0e73746b.png
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1248467511, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TMWHN6S5YD5BOUXAHLV6NUTXANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
— Reply to this email directly, view it on GitHubhttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fmaximilianh%2FcellBrowser%2Fissues%2F251%23issuecomment-1248617965&data=05%7C01%7Ctkakar%40wpi.edu%7Cbaafbb1a637e4f93cc9e08da975c4b69%7C589c76f5ca1541f9884b55ec15a0672a%7C0%7C0%7C637988719931785115%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=FhWb%2FAB71jj3M5P2SD7fQKjiYBvSRnO%2B3o%2F9XJzdVXE%3D&reserved=0, or unsubscribehttps://nam11.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.com%2Fnotifications%2Funsubscribe-auth%2FADLQKX33F767Z6AAGXH4T7TV6OEDJANCNFSM6AAAAAAQKKDPJQ&data=05%7C01%7Ctkakar%40wpi.edu%7Cbaafbb1a637e4f93cc9e08da975c4b69%7C589c76f5ca1541f9884b55ec15a0672a%7C0%7C0%7C637988719931785115%7CUnknown%7CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0%3D%7C3000%7C%7C%7C&sdata=HmMOjaFb0PczDymKe8w42sAa8C4H%2FUU511aelqsRPf8%3D&reserved=0. You are receiving this because you commented.Message ID: @.***>
Hi @tkakar and @Bhersbach, sorry that it took me so long to understand what's going on. It had nothing to do with the hierarchy, but was a typo in the new diseases filter, I had fixed that in the develop branch, it slipped into the last release. Here is a new pre-release https://pypi.org/project/cellbrowser/1.2.1a1/ (I must release soon) You can install it with "pip install --pre cellbrowser --upgrade"
@maximilianh thanks for fixing the issue. I did as suggested. The disease
related error is gone.
The individual datasets can be accessed via ?ds=datasetName
on the url, but the data hierarchy still does not work. Under developer tools, I see errors. Posting screen shot of both scenarios (individual dataset and hiararchy).
Hmm....It would be so much easier to debug this if you had this on a public website so I could run the debugger instead of guessing...
I have a feeling that we've fixed this bug in the develop branch...
Can you send me the code around line 1138 in your version cellBrowser.js ? Or send the me the whole file? Somehow, in my cellBrowser.js in the master branch, it doesn't match what your error message shows.
Don't you want to make the website publicly accessible? You'll have to do this anyways sat some point, right? If you want to keep the private, you can send me the URL via direct email.
On Fri, Sep 16, 2022 at 4:54 PM Tabassum Kakar @.***> wrote:
@maximilianh https://github.com/maximilianh thanks for fixing the issue. I did as suggested. The disease related error is gone. The individual datasets can be accessed via ?ds=datasetName on the url, but the data hierarchy still does not work. Under developer tools, I see errors. Posting screen shot of both scenarios (individual dataset and hiararchy). [image: Screen Shot 2022-09-16 at 10 49 42 AM] https://user-images.githubusercontent.com/14091615/190668367-485e6242-d24d-493b-b673-a55caecae6b3.png [image: Screen Shot 2022-09-16 at 10 49 10 AM] https://user-images.githubusercontent.com/14091615/190668372-cdfbf1ad-38fd-4d76-b4f2-89ccaf6358ff.png
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1249470117, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TP5E5UMH7F54U3QUGDV6SC3FANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you were mentioned.Message ID: @.***>
@maximilianh. I'm emailing the whole file, but the code around line 1138 is below
for (var category of categories) {
var filtList = filtVals[category];
var domList = domVals[category];
let found = false;
if (filtList.length===0)
found = true;
for (var filtVal of filtList)
if (domList.indexOf(filtVal)!=-1)
found = true;
if (!found) {
isShown = false;
break;
}
}
Unfortunately, due to data confidentiality, I cannot share the website.
Sorry, again a question: which line here is line 1138 ?
On Mon, Sep 19, 2022 at 6:16 PM Tabassum Kakar @.***> wrote:
@maximilianh https://github.com/maximilianh. I'm emailing the whole file, but the code around line 1138 is below `for (var category of categories) { var filtList = filtVals[category]; var domList = domVals[category];
let found = false; if (filtList.length===0) found = true; for (var filtVal of filtList) if (domList.indexOf(filtVal)!=-1) found = true; if (!found) { isShown = false; break; } }`
Unfortunately, due to data confidentiality, I cannot share the website.
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1251238940, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TP2QNSCL26YXXR7BTDV7CGU5ANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you were mentioned.Message ID: @.***>
It's this line. if (filtList.length===0)
So I'm sure that you got the hierarchy to work, now it's something with the filters that is broken on your installation. I wonder if you specified some filters but not others... or the code has a problem (that is not visible on our site).
OK, I took the entire develop branch and made a pip release from it. Can
you do a "pip install cellbrowser --pre --upgrade" and then run "cbUpgrade
-o
If it's not fixed, then it would be helpful to see your cellbrowser.conf files. Or - if you know how to do that - you could put a breakpoint at line 1138 and let me know what the value of the variable "category" is. Or you could add console.log(category) before line 1138 and send us a copy of the console.
This will be fixed one way or another, thanks for bearing with me. It's not easy to debug without seeing the website nor the data.
On Mon, Sep 19, 2022 at 6:22 PM Tabassum Kakar @.***> wrote:
It's this line. if (filtList.length===0)
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1251247107, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TNV67EZXPCKSUKZAV3V7CHNXANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you were mentioned.Message ID: @.***>
@maximilianh thanks for troubleshooting this with me. My .conf files were based on some previous version from repo where other categories were not defined (except the body_parts). I just updated it and it's working now. Thanks for all your help.
Ahhh. This makes sense. So something is broken when these are not specified. I guess we never tested this.
@Brittney Wick @.***> don't we have a dataset on cells-test where some of these attributes are missing? Hmmm... I'm surprised this hasn't come up before...
On Tue, Sep 20, 2022 at 4:04 PM Tabassum Kakar @.***> wrote:
@maximilianh https://github.com/maximilianh thanks for troubleshooting this with me. My .conf files were based on some previous version from repo where other categories were not defined (except the body_parts). I just updated it and it's working now. Thanks for all your help.
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1252404873, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TNSG37HCU7PC2AXSUDV7HAAFANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you were mentioned.Message ID: @.***>
Yes, please don't use the -r option, it's not meant to be used with the -p option, sorry I didn't mention that (and it's not documented I just saw).
No need for the -r just run cbBuild with the -o and -p options.
Also, if this is a production webserver, don't use the -p built-in server, it's slow and can use a lot of RAM. I'd rather use any other webserver, apache, nginx or anything else you like.
On Mon, Sep 12, 2022 at 4:29 PM Bhersbach @.***> wrote:
Done, but to no avail. Also the command cbBuild -r -o /home/cellbrowser/html -p 8999 runs but does not yields a message like: Serving /home/cellbrowser/html/ on port 8999 Point your internet browser to http://127.0.0.1:8999 (or the IP address of this server)
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1243829271, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TKLQMH7XFDYII3ETB3V5446LANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
Oops, sorry! I should have written "desc.conf", in addition to "cellbrowser.conf". You define the title in cellbrowser.conf like this /home/cellbrowser/User1/:
shortLabel = "All Datasets"
and then write your html into desc.conf, as you would do for a normal dataset:
title = "UCSC Cell Browser Intro" abstractFile="../news/combined.html"
You can create sample files for the root dataset as with any other dataset:
cd /home/cellbrowser/User1/ cbBuild --init
If you have ideas on how we could better document this or make it easier, let me know...
On Mon, Sep 12, 2022 at 3:42 PM Bhersbach @.***> wrote:
I hadn't done that and just gave it a try, it doesn't seem to have solved the issue (or I did it incorrectly). Unfortunately, our instance is currently not public, so I will try to briefly summarize the situation:
- We have one folder (User1) that contains two datasets (Dataset1 & Dataset2)
- The ~/.cellbrowser.conf file does not seem to be present the home folder so I ran export CBDATAROOT='/home/cellbrowser/User1/'
- I copied cellbrowser.conf and desc.conf from Dataset1 to the User1 folder and edited as described in the link
- I run cbBuild -r /home/cellbrowser/html -p 8999 within the User1 folder, this runs but does not start a CellBrowser server
- I tried running the same command within the Dataset1 folder, which runs and creates a CellBrowser server but no index/title page
— Reply to this email directly, view it on GitHub https://github.com/maximilianh/cellBrowser/issues/251#issuecomment-1243763380, or unsubscribe https://github.com/notifications/unsubscribe-auth/AACL4TJ5FSF55MF63RFMO33V54XLZANCNFSM6AAAAAAQKKDPJQ . You are receiving this because you commented.Message ID: @.***>
We are setting up Cell Browser in our lab and have successfully prepared two Cell Browser folders as a test. However, when we use cbBuild to host the data, we can access the individual datasets only by using the dataset name, i.e., server address:/?ds=dataset_name. The index/title page, does not load and, instead, we get an empty Cell Browser page. cbBuild was ran with the same output folder and indeed confirms it found two datasets within that folder. When attempting to open a dataset through file > open dataset, no options are given either. Is there something we are forgetting to specify?
System specs: OS: Linux (Ubuntu 18.04) Cell Browser v1.2.1