Iancu, O. | bioRxiv Science

bioRxiv2026

Face ensembles reshape the neural other-race effect

The other-race effect (ORE), poorer recognition of other-race (OR) than same-race (SR) faces, is established for individual faces, but its neural expression during group viewing remains unclear. Twenty-two East Asian adults viewed East Asian and White faces in different formats: individually and in six-face ensembles during EEG recording. Behavioral testing confirmed an SR advantage. Decoding and generative reconstruction characterized neural discriminability, geometry, dynamics, sensor-level information, and recoverable content. Single SR faces were more discriminable, more dispersed in face space, and reconstructed more accurately. Ensembles preserved racial-composition information but reduced the cumulative SR-OR decoding difference and showed instead an early OR advantage. Reconstructions supported identification of individual faces and ensemble summaries, with an overall SR advantage but no reliable race-by-format interaction. Independent judgments revealed race- and format-dependent shifts in reconstructed age, valence, and arousal. These findings reveal a neural ORE whose expression depends on viewing context and representational measure.

neuroscience↗

bioRxiv2022

Multiplex HDR for Disease and Correction Modeling of SCID by CRISPR Genome Editing in Human HSPCs

Severe combined immunodeficiency (SCID) is a group of monogenic primary immunodeficiencies caused by mutations in genes involved in the process of lymphocyte maturation and function. CRISPR-Cas9 gene editing of the patients own hematopoietic stem and progenitor cells (HSPCs) ex vivo could provide a therapeutic alternative to allogeneic hematopoietic stem cell transplantation (HSCT), the current gold standard for treatment of SCID. Using CRISPR-Cas9/rAAV6 gene-editing, we engineered genotypes in healthy donor (HD)-derived CD34+ HSPCs, thus eliminating the need for rare patient samples, to model both SCID and the therapeutic outcomes of gene-editing therapies for SCID via multiplexed homology directed repair (HDR). Firstly, we developed a SCID disease model via knock-out of both alleles of genes critical to the development of lymphocytes; and secondly, we established a knock-in/knock-out (KI-KO) strategy to develop a proof-of-concept single-allelic gene correction. Since SCID is a recessive disorder, correction of only one allele is enough to cure the patient. Based on these results, we performed gene correction of RAG2-SCID patient-derived CD34+ HSPCs that successfully developed into CD3+ T cells with diverse TCR repertoires in an in vitro T-cell differentiation (IVTD) platform. By using CRISPR-Cas9, multiplexed HDR, HD-derived CD34+ HSPCs, and an IVTD system we outline an approach for the study of human lymphopoiesis. We present both a way for researchers to determine the optimal configuration for CRISPR-Cas9 gene correction of SCID and other recessive blood disorders, and the feasibility of translating these techniques to perform gene correction in patient-derived CD34+ HSPCs.

bioengineering↗