Clemens, B. | Search bioRxiv

bioRxiv2022

Accurate sex prediction of cisgender and transgender individuals without brain size bias

Brain size differs substantially between human males and females. This difference in total intracranial volume (TIV) can cause bias when employing machine-learning approaches for the investigation of sex differences in brain morphology. TIV-biased models will likely not capture actual qualitative sex differences in brain organization but rather learn to classify an individuals sex based on brain size differences, thus leading to spurious and misleading conclusions, for example when comparing brain morphology between cisgender- and transgender individuals. Here, TIV bias in sex classification models applied to cis- and transgender individuals was systematically investigated by controlling for brain size either through featurewise confound removal or by matching training samples for TIV. Our results provide evidence that non-TIV-biased models can classify the sex of both cis- and transgender individuals with high accuracy, highlighting the importance of appropriate modelling to avoid bias in automated decision making. TeaserAccurate non-biased structural sex classification in cis- and transgender individuals by matching training samples for TIV

neuroscience↗

bioRxiv2018

A Multiplex Droplet Digital PCR Assay for Quantification of HTLV-1c DNA Proviral Load and T-Cells from Blood and Respiratory Exudates Sampled in a Remote Setting

During human T-cell leukemia virus type-1 (HTLV-1) infection the frequency of cells harboring an integrated copy of viral cDNA, the proviral load (PVL), is the main risk factor for progression of HTLV-1-associated diseases. Accurate quantification of provirus by droplet digital PCR (ddPCR) is a powerful diagnostic tool with emerging uses for monitoring viral expression. Current ddPCR techniques quantify HTLV-1 PVL in terms of whole genomic cellular material, while the main target of HTLV-1 infection is the CD4+ and CD8+ T-cell. Our understanding of HTLV-1 proliferation and the amount of viral burden present in different compartments is limited. Recently a sensitive ddPCR assay was applied to quantifying T-cells by measuring loss of germline T-cell receptor genes as method of distinguishing non-T-cell from recombined T-cell DNA. In this study, we demonstrated and validated novel applications of the duplex ddPCR assay to quantify T-cells from various sources of human gDNA extracted from frozen material (PBMCs, bronchoalveolar lavage, and induced sputum) from a cohort of remote Indigenous Australians and then compared the T-cell measurements by ddPCR to the prevailing standard method of flow cytometry. The HTLV-1c PVL was then calculated in terms of extracted T-cell gDNA from various compartments. Because HTLV-1c preferentially infects CD4+ T-cells, and the amount of viral burden correlates with HTLV-1c disease pathogenesis, application of this ddPCR assay to accurately measure HTLV-1c-infected T-cells can be of greater importance for clinical diagnostics, prognostics as well as monitoring therapeutic applications.

microbiology↗