s3seg fails on single-channel images

[ArtemSokolov]

• Download exemplar-001:

curl -f -o test.ome.tif https://mcmicro.s3.amazonaws.com/ci/exemplar-001-cycle6.ome.tif

• Cut the image down to the first channel (recyze uses 0-based channel indexing):

nextflow run labsyspharm/mcmicro/roadie.nf --do recyze --in test.ome.tif --channels 0

• Verify image dimensionality:

$ python -c "import tifffile; print(tifffile.TiffFile('test.ome.tif').series[0].shape)"
(4, 3139, 2511)

$ python -c "import tifffile; print(tifffile.TiffFile('test_crop.ome.tif').series[0].shape)"
(3139, 2511)

• UnMicst runs fine on the single-channel image:

docker run --rm -v "$PWD":/data labsyspharm/unmicst:2.7.4 python /app/unmicstWrapper.py /data/test_crop.ome.tif --stackOutput --outputPath /data --channel 1

• However, s3seg fails:

docker run --rm -v "$PWD":/data labsyspharm/s3segmenter:1.3.18 python /app/S3segmenter.py --imagePath /data/test_crop.ome.tif --stackProbPath /data/test_crop_Probabilities_1.tif --probMapChan 1 --outputPath /data

with an out-of-bounds errors:

Traceback (most recent call last):
  File "/app/S3segmenter.py", line 484, in <module>
    tissueCrop =tifffile.imread(imagePath,key=iChan)
  File "/usr/local/lib/python3.8/site-packages/tifffile/tifffile.py", line 800, in imread
    return tif.asarray(**kwargs)
  File "/usr/local/lib/python3.8/site-packages/tifffile/tifffile.py", line 3128, in asarray
    pages = self.pages._getlist(key)
  File "/usr/local/lib/python3.8/site-packages/tifffile/tifffile.py", line 5311, in _getlist
    return [getitem(key)]
  File "/usr/local/lib/python3.8/site-packages/tifffile/tifffile.py", line 5360, in _getitem
    self._seek(key)
  File "/usr/local/lib/python3.8/site-packages/tifffile/tifffile.py", line 5281, in _seek
    raise IndexError('index out of range')
IndexError: index out of range

[AlbertPlaPlanas]

@ArtemSokolov did you manage to find a solution for this error? We are experiencing the exact same issue.

[clarenceyapp]

S3segmenter by default uses a background or autofluorescence channel to detect the tissue bounds. If you don't have such a channel, the TissueMaskChan parameter should be set to 1 so that it does not consider the cytoplasmic regions. CytoMaskChan similarly assumes a separate channel index. Clarence

…

________________________________ From: AlbertPlaPlanas ***@***.***> Sent: Tuesday, October 24, 2023 8:00:47 AM To: HMS-IDAC/S3segmenter ***@***.***> Cc: Subscribed ***@***.***> Subject: Re: [HMS-IDAC/S3segmenter] s3seg fails on single-channel images (Issue #16) @ArtemSokolov<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_ArtemSokolov&d=DwMFaQ&c=WO-RGvefibhHBZq3fL85hQ&r=Hv2ImNU5v_QaacHDz01XFRrD_bA8Sg7kSTOh8SrzhDk&m=4P3KJEE_sr0xk_CUSZxy8XmcQyfRqTLM3Yq4O3sS1UVBBcVCt1f5F9yqYm5JVR7f&s=BBKOm7K76CSZA02oXuGMMFinAc5Y3XMRZXagb9o5cGQ&e=> did you manage to find a solution for this error? We are experiencing the exact same issue. — Reply to this email directly, view it on GitHub<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_HMS-2DIDAC_S3segmenter_issues_16-23issuecomment-2D1777070554&d=DwMFaQ&c=WO-RGvefibhHBZq3fL85hQ&r=Hv2ImNU5v_QaacHDz01XFRrD_bA8Sg7kSTOh8SrzhDk&m=4P3KJEE_sr0xk_CUSZxy8XmcQyfRqTLM3Yq4O3sS1UVBBcVCt1f5F9yqYm5JVR7f&s=FQFUgdzy0HTjIM3oyXd4L6ZsBBzJE_yfA_UBEhO4rV4&e=>, or unsubscribe<https://urldefense.proofpoint.com/v2/url?u=https-3A__github.com_notifications_unsubscribe-2Dauth_AFLKVSYHKDBWK2TWVZQYQEDYA6UW7AVCNFSM6AAAAAAQIE7IHWVHI2DSMVQWIX3LMV43OSLTON2WKQ3PNVWWK3TUHMYTONZXGA3TANJVGQ&d=DwMFaQ&c=WO-RGvefibhHBZq3fL85hQ&r=Hv2ImNU5v_QaacHDz01XFRrD_bA8Sg7kSTOh8SrzhDk&m=4P3KJEE_sr0xk_CUSZxy8XmcQyfRqTLM3Yq4O3sS1UVBBcVCt1f5F9yqYm5JVR7f&s=CSucPWGBNGQZGpodpESMkIXTHP1uGJllib2clrn5S5U&e=>. You are receiving this because you are subscribed to this thread.Message ID: ***@***.***>

[ArtemSokolov]

@AlbertPlaPlanas If you are running this through MCMICRO, adding the following to params.yml should address the issue:

options:
  s3seg: --TissueMaskChan 1

@clarenceyapp Would it make sense to make 1 be the default for that argument?