MapMyCells Down?

How To MapMyCells
1

Hello!

I am having problems using the MapMyCells site. My goal is to map some TRAP-seq data as a test. I always get this notification.

Screenshot 2026-03-09 at 12.09.15 PM
Screenshot 2026-03-09 at 12.09.15 PM1360×420 60.5 KB

Apologies if I am missing something I need to do to be able to use the the site. Any help would be appreciated.

Best,

Calvin Liu

2

Hi,

I’m sorry about that.

I was just (09:26 Pacific Daylight Time/ 16:26 UTC; about 15 minutes after you posted your message) able to run a mapping job. Do you mind submitting your job again to see if this is a transient bug that has resolved itself? If you encounter the same problem, I will start a conversation with the engineers who might be able to provide insight into this kind of failure.

Thanks,

Scott

3

Hello Scott,

Thank you so much for your quick response no need to be sorry! This is currently what I see.

Screenshot 2026-03-09 at 2.00.48 PM
Screenshot 2026-03-09 at 2.00.48 PM3420×1830 389 KB

I still got the same data error and after it disappears this is whats left.

I actually did a small test. And I used a personal cellular hotspot instead of my companies wifi and it is now loading and working.

Best,

Calvin Liu

4

Okay. Interesting. I know a lot about what the mapping code does but significantly less about the infrastructure that gets data into the mapping code. I will ask the relevant team over here if there is, for instance, a known firewall configuration that could be preventing you from accessing MapMyCells over your company wifi.

Does your company of a particularly restrictive firewall in place?

The error you are seeing generally means that MapMyCells is unable to load the data it needs to know which taxonomies are available for mapping, which is not a problem I have encountered before.

5

Hello Scott,

Im currently in a startup incubator so I would not be surprised if there is restrictions in place that prevented me from using MapMyCells.

I hope it’s okay to post here, happy to create a separate forum post if preferred!

I am working with bulk TRAP-seq data and am trying to use MapMyCells for cell type mapping. I wanted to ask for advice on the best approach given the nature of this data type.

A bit of background on my data:

  • Data is already normalized as TPM (Transcripts Per Million) per gene
  • I have both bound (cell-type enriched translatome) and unbound (bulk background) fractions
  • I am working with a known cell population
  • I currently have one sample I am trying to map

My current approach:

  1. Extracting the gene column and the bound sample TPM values
  2. Transposing to cells (rows) x genes (columns) format
  3. Approximating CP10K normalization via TPM/100, then applying log1p transform
  4. Saving as h5ad and uploading to MapMyCells

My concerns and questions:

  1. Is TPM/100 → log1p a reasonable approximation for MapMyCells input, given I no longer have access to raw counts?
6

(this thread is probably fine)

If I interpret what you are saying correctly, your data should be fine as-is.

You say that your data is in TPM, which I interpret to mean that the count vector for each cell is multiplied by whatever factor is needed to get the vector to sum to 1 million.

The first thing that MapMyCells will do is convert your data to log2(CPM+1) so it can compare your data to the reference data, which is also stored in log2(CPM+1) (“CPM” = “counts per million”, which I assume is equivalent to “transcripts per million”). So, what MapMyCells will do is

for each cell

  1. multiply the expression vector by whatever factor gets the vector to sum to 1 million
  2. add one
  3. take the log base 2

There should be no consequence to applying step (1) twice. The only problem would arise if you took the logarithm twice. So, again, if I understood you correctly, I would say: go ahead and submit your data without doing any transformations to it.