Immunized Patients Face Reduced Access to Transplantation in the Eurotransplant Kidney Allocation System.

BACKGROUND: The presence of donor-specific HLA antibodies before transplantation is associated with poor transplantation outcomes. Unacceptable antigens can be assigned for Eurotransplant kidney transplant candidates to prevent kidney offers against which the candidate has developed clinically relevant HLA antibodies. This retrospective cohort study aimed to assess to what degree unacceptable antigens affect access to transplantation in the Eurotransplant Kidney Allocation System (ETKAS). METHODS: Candidates who underwent kidney-only transplantation between 2016 and 2020 were included (n = 19 240). Cox regression was used to quantify the relationship between the relative transplantation rate and virtual panel-reactive antibodies (vPRAs), which is the percentage of the donor pool with unacceptable antigens. Models used accrued dialysis time as the timescale; were stratified by country and blood group of patient and were adjusted for nontransplantable status, patient age, sex, history of kidney transplantations, and prevalence of 0 HLA-DR-mismatched donors. RESULTS: Transplantation rates were 23% lower for vPRA 0.1% to 50%, 51% lower for vPRA 75% to 85%, and decreased rapidly for vPRA of >85%. Prior studies showed significantly lower ETKAS transplantation rates only for highly sensitized patients (vPRA of >85%). The inverse relationship between transplantation rate and vPRA is independent of Eurotransplant country, listing time, and 0 HLA-DR-mismatched donor availability. Results were similar when quantifying the relationship between vPRA and attainment of a sufficiently high rank for an ETKAS offer, suggesting lower transplantation rates for immunized patients are due to current ETKAS allocation. CONCLUSIONS: Immunized patients face lower transplantation rates across Eurotransplant. The current ETKAS allocation mechanism inadequately compensates immunized patients for reduced access to transplantation.

PIPENN: protein interface prediction from sequence with an ensemble of neural nets.

MOTIVATION: The interactions between proteins and other molecules are essential to many biological and cellular processes. Experimental identification of interface residues is a time-consuming, costly and challenging task, while protein sequence data are ubiquitous. Consequently, many computational and machine learning approaches have been developed over the years to predict such interface residues from sequence. However, the effectiveness of different Deep Learning (DL) architectures and learning strategies for protein-protein, protein-nucleotide and protein-small molecule interface prediction has not yet been investigated in great detail. Therefore, we here explore the prediction of protein interface residues using six DL architectures and various learning strategies with sequence-derived input features. RESULTS: We constructed a large dataset dubbed BioDL, comprising protein-protein interactions from the PDB, and DNA/RNA and small molecule interactions from the BioLip database. We also constructed six DL architectures, and evaluated them on the BioDL benchmarks. This shows that no single architecture performs best on all instances. An ensemble architecture, which combines all six architectures, does consistently achieve peak prediction accuracy. We confirmed these results on the published benchmark set by Zhang and Kurgan (ZK448), and on our own existing curated homo- and heteromeric protein interaction dataset. Our PIPENN sequence-based ensemble predictor outperforms current state-of-the-art sequence-based protein interface predictors on ZK448 on all interaction types, achieving an AUC-ROC of 0.718 for protein-protein, 0.823 for protein-nucleotide and 0.842 for protein-small molecule. AVAILABILITY AND IMPLEMENTATION: Source code and datasets are available at https://github.com/ibivu/pipenn/. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.