modified visium read function for reading tissue_positions.csv properly

[shashkat]

Description

Modified visium read function (squidpy.read.visium) by correcting header argument of pd.read_csv when reading tissue_positions.csv (Ref: #1089). Earlier, it was reading the tissue_positions.csv with pd.read_csv(file, header=1, ...) which was leading to the second row of the csv being read as header, which skipped one row of data. Changing to header=0 makes it function as expected. I don't know why it was header=1 in the first place (my guess is likely there used to be an extra line in tissue_positions.csv earlier, which is not there in the outputs now):

$ head tissue_positions.csv
barcode,in_tissue,array_row,array_col,pxl_row_in_fullres,pxl_col_in_fullres
GTCACTTCCTTCTAGA-1,1,0,0,23891.771232682222,39028.97429208872
CACGGTCTCCTTACGA-1,1,0,2,23897.618791707442,38663.634433938605
ATAGCTGCGGATAAGA-1,1,0,4,23903.46462886722,38298.2952273851
GTCAGTATGTCCGGCG-1,1,0,6,23909.31218148734,37932.95571104603
ATGTACCAGTTACTCG-1,1,0,8,23915.159740512565,37567.615852895906
ACGCTCAGTGCACCGT-1,1,0,10,23921.00558727999,37202.276133625826
TCACTAACGTATAGTT-1,1,0,12,23926.853130292468,36836.93713000333
CGGTTAGGCCTGGACG-1,1,0,14,23932.699492979013,36471.59724943537
GATATCACCAGCATGG-1,1,0,16,23938.546539287658,36106.257381677606

Closes

Closes #1089

[codecov]

Codecov Report

✅ All modified and coverable lines are covered by tests.
✅ Project coverage is 66.17%. Comparing base (83a11d5) to head (06af281).

Additional details and impacted files

@@           Coverage Diff           @@
##             main    #1094   +/-   ##
=======================================
  Coverage   66.17%   66.17%           
=======================================
  Files          44       44           
  Lines        7160     7160           
  Branches     1218     1218           
=======================================
  Hits         4738     4738           
  Misses       1944     1944           
  Partials      478      478           

Files with missing lines	Coverage Δ
src/squidpy/read/_read.py	35.16% <ø> (ø)

🚀 New features to boost your workflow:

• ❄️ Test Analytics: Detect flaky tests, report on failures, and find test suite problems.

[selmanozleyen]

thanks a lot, have you tried this with all the notebooks? When I check the notebooks and make sure the case when we need header=1 is more clear. I'd like to clarify the reason why it was written like that initially @timtreis @giovp any ideas?

The line was written by @stephenwilliams22 and @michalk8 , in #583.

Do you guys remember the reason for this? in which case would it crash if we set header=0

[shashkat]

Hello, thanks for your points. I have tested on some notebooks in docs/notebooks/examples, and it seems to work as expected. I will check on the other notebooks too in a few days and confirm if it looks fine.

Thanks!

[shashkat]

Hello, I ran all the notebooks in docs/notebooks/examples and relevant notebooks in docs/notebooks/tutorials and they seem to run successfully with the change, and the outputs seem to match visually at least. Let me know if anything else is required!