Multiplex Slides

Multiplex slides store independent biomarker or channel data. ASlide exposes them through explicit channel reads instead of generic RGB reads.

Family Semantics

For multiplex slides:

slide.slide_family == "multiplex"

Multiplex slides support list_biomarkers(), get_default_display_biomarker(), and read_biomarker_region(). Generic read_region() is rejected so callers must choose the intended channel.

Biomarker Reads

from Aslide import Slide

with Slide("sample.qptiff") as slide:
    biomarkers = slide.list_biomarkers()
    marker = biomarkers[0]
    region = slide.read_biomarker_region((0, 0), 0, (512, 512), marker)

read_biomarker_region(location, level, size, biomarker) returns an RGBA image. slide.supports_biomarkers is true for multiplex slides even when static registry capabilities are conservative for runtime-classified formats.

Default Display Biomarker

Some backends define a default display channel. QPTIFF commonly uses DAPI as the default display biomarker. IMS and other multiplex backends may derive a default from channel metadata.

with Slide("sample.ims") as slide:
    marker = slide.get_default_display_biomarker()
    region = slide.read_biomarker_region((0, 0), 0, (512, 512), marker)

If no stable default exists, ASlide raises MissingDefaultBiomarkerError.

Runtime-Classified Formats

QPTIFF and CZI are classified after the file is opened. A brightfield QPTIFF behaves like an ordinary brightfield slide. A multiplex QPTIFF requires biomarker-aware reads.

with Slide("sample.qptiff") as slide:
    if slide.slide_family == "brightfield":
        region = slide.read_region((0, 0), 0, (512, 512))
    else:
        marker = slide.get_default_display_biomarker()
        region = slide.read_biomarker_region((0, 0), 0, (512, 512), marker)

CZI support can use the lightweight czi extra or the Bio-Formats path. Metadata determines whether the opened file behaves as brightfield or multiplex.

TIFF, HDF5, H5AD, IMS, and MCD

Opening an OME-like TIFF anchor file can resolve to a multiplex backend when the file contains compatible marker/channel data. Generic .tif and .tiff files still fall back to ordinary TIFF behavior when they do not match the multiplex probe.

Image-backed .h5, .hdf5, and .h5ad containers resolve to multiplex behavior when they expose a direct channel-by-height-by-width raster plus marker metadata. Table-only AnnData .h5ad files are not reconstructed into slides by this path.

IMS files use a dedicated multiplex backend that reads channel metadata and pyramid levels directly from the IMS container.

MCD files use readimc. When an MCD file contains multiple acquisitions, ASlide selects the largest acquisition by pixel area by default and stores selected acquisition metadata in slide.properties. Pass acquisition_id=<id> to choose a specific acquisition.

with Slide("sample.mcd", acquisition_id=2) as slide:
    print(slide.properties.get("mcd.selected-acquisition-description"))
    print(slide.list_biomarkers()[:5])