Search bioRxiv⌕ Search

Biology subjects

Harrizi, S.

Publications and source records attributed to Harrizi, S..

2 recordsLinked to original sources

Gene identity, not variant effect, dominates ClinVar benchmarks of missense pathogenicity predictors

Missense pathogenicity predictors are routinely benchmarked against ClinVar, whose labels are strongly structured by gene: genes under diagnostic scrutiny accumulate pathogenic submissions while incidentally sequenced genes accumulate benign ones. We asked how much of a benchmark score this structure alone can produce. On 197,904 ClinVar missense variants validated against UniProt canonical sequences, a null model using no variant-level information, scoring each variant only by the pathogenic fraction of its own gene, reaches an area under the receiver operating characteristic curve (AUROC) of 0.921 under a random 10-fold split. On a common intersection of 169,989 variants, four current predictors exceed it by only 0.036 to 0.044. The inflation is not uniform, so it does not cancel when predictors are compared: under within-gene evaluation the ranking inverts, AlphaMissense rising from third to first and gMVP falling to third (p < 0.0001). The inversion survives removal of ceiling genes and replicates on an independently curated benchmark. Because both rankings derive from the same ClinVar labels, we arbitrated between them using data with no gene-level structure: agreement with 47 human deep mutational scanning assays matches the within-gene ranking and inverts the conventional one (p = 0.027, 0.0023). Across twenty-two dbNSFP predictors scored on one common intersection of 112,248 variants, with each tool's exposure to clinical labels registered before any score was extracted, predictors never trained on such labels sit 0.051 AUROC behind supervised ones globally but only 0.026 behind within genes (difference +0.025 [+0.023, +0.027], p < 0.0001). Leave-one-out correction, the standard remedy, is worth 0.002 AUROC. Much of ClinVar benchmark performance reflects gene identity rather than variant effect, and the distortion changes which predictor a benchmark ranks first, in a direction experimental data contradicts. We release genenull, a single-file implementation, so reporting this baseline costs one function call.

bioinformatics↗

A distinctive tumor compartment in pancreatic lobules defined by nascent stroma and classical tumor cell phenotype

Pancreatic ductal adenocarcinoma (PDAC) is a highly aggressive tumor type characterized by a particularly extensive stroma. While different types of cancer-associated fibroblasts (CAFs) in this desmoplastic stroma have been described, areas of early invasion and nascent stroma are understudied. Here, we identify a distinctive PDAC niche within the pancreatic lobules, a compartment dominated by pancreatic exocrine cells and slender stroma. Cellular interaction profiling using machine learning on whole slide images of human PDAC reveals that the tumor invasion front in the lobules is dominated by specific interactions of tumor cells and exocrine cells that have undergone acinar-to-ductal metaplasia (ADM). Multiplex protein and mRNA stains confirm that tumor growth in the lobules is closely linked to ADM in the lobules, and reveal stromal protein gradients from the gracile lobular stroma to the characteristic desmoplastic stroma. We identify nascent CAFs (nCAFs), co-expressing expressing nerve growth factor receptor (NGFR) and platelet-derived growth factor receptor alpha (PDGFRa) that are absent in the mature, desmoplastic stroma. Lobular invasion and nCAFs are intertwined with phenotypic changes of the cancer cells, such that tumor cells in lobules express classical subtype markers, while those embedded in the desmoplastic are on the basal end of the phenotypic continuum. In mice, the PDAC subtype - basal or classical - similarly depends on tissue location, suggesting microenvironmental factors rather than clonal selection as important drivers of tumor phenotype identity. Clinically, our results mandate factoring in tumor tissue location when calling PDAC subtypes. Biologically, they identify pancreatic lobules as a distinctive tissue niche associated with nascent stroma, and they suggest that lobular colonization by tumor cells is a significant route of PDAC progression.

cancer biology↗