RESUMEN
An open challenge in human genetics is to better understand the systems-level impact of genotype variation on developmental cognition. To characterize the genetic underpinnings of peri-adolescent cognition, we performed genotype-phenotype and systems analysis for binarized accuracy in nine cognitive tasks from the Philadelphia Neurodevelopmental Cohort (~2,200 individuals of European continental ancestry aged 8-21 years). We report a region of genome-wide significance within the 3' end of the Fibulin-1 gene (P = 4.6 × 10-8), associated with accuracy in nonverbal reasoning, a heritable form of complex reasoning ability. Diffusion tensor imaging data from a subset of these participants identified a significant association of white matter fractional anisotropy with FBLN1 genotypes (P < 0.025); poor performers show an increase in the C and A allele for rs77601382 and rs5765534, respectively, which is associated with increased fractional anisotropy. Integration of published human brain-specific 'omic maps, including single-cell transcriptomes of the developing human brain, shows that FBLN1 demonstrates greatest expression in the fetal brain, as a marker of intermediate progenitor cells, demonstrates negligible expression in the adolescent and adult human brain, and demonstrates increased expression in the brain in schizophrenia. Collectively these findings warrant further study of this gene and genetic locus in cognition, neurodevelopment, and disease. Separately, genotype-pathway analysis identified an enrichment of variants associated with working memory accuracy in pathways related to development and to autonomic nervous system dysfunction. Top-ranking pathway genes include those genetically associated with diseases with working memory deficits, such as schizophrenia and Parkinson's disease. This work advances the "molecules-to-behavior" view of cognition and provides a framework for using systems-level organization of data for other biomedical domains.
Asunto(s)
Imagen de Difusión Tensora , Sustancia Blanca , Adulto , Humanos , Adolescente , Cognición/fisiología , Encéfalo/diagnóstico por imagen , Encéfalo/fisiología , GenómicaRESUMEN
OBJECTIVES: Previous trial results suggest that only a small number of patients with non-metastatic renal cell carcinoma (RCC) benefit from adjuvant therapy. We assessed whether the addition of CT-based radiomics to established clinico-pathological biomarkers improves recurrence risk prediction for adjuvant treatment decisions. METHODS: This retrospective study included 453 patients with non-metastatic RCC undergoing nephrectomy. Cox models were trained to predict disease-free survival (DFS) using post-operative biomarkers (age, stage, tumor size and grade) with and without radiomics selected on pre-operative CT. Models were assessed using C-statistic, calibration, and decision curve analyses (repeated tenfold cross-validation). RESULTS: At multivariable analysis, one of four selected radiomic features (wavelet-HHL_glcm_ClusterShade) was prognostic for DFS with an adjusted hazard ratio (HR) of 0.44 (p = 0.02), along with American Joint Committee on Cancer (AJCC) stage group (III versus I, HR 2.90; p = 0.002), grade 4 (versus grade 1, HR 8.90; p = 0.001), age (per 10 years HR 1.29; p = 0.03), and tumor size (per cm HR 1.13; p = 0.003). The discriminatory ability of the combined clinical-radiomic model (C = 0.80) was superior to that of the clinical model (C = 0.78; p < 0.001). Decision curve analysis revealed a net benefit of the combined model when used for adjuvant treatment decisions. At an exemplary threshold probability of ≥ 25% for disease recurrence within 5 years, using the combined versus the clinical model was equivalent to treating 9 additional patients (per 1000 assessed) who would recur without treatment (i.e., true-positive predictions) with no increase in false-positive predictions. CONCLUSION: Adding CT-based radiomic features to established prognostic biomarkers improved post-operative recurrence risk assessment in our internal validation study and may help guide decisions regarding adjuvant therapy. KEY POINTS: In patients with non-metastatic renal cell carcinoma undergoing nephrectomy, CT-based radiomics combined with established clinical and pathological biomarkers improved recurrence risk assessment. Compared to a clinical base model, the combined risk model enabled superior clinical utility if used to guide decisions on adjuvant treatment.
Asunto(s)
Carcinoma de Células Renales , Neoplasias Renales , Humanos , Niño , Carcinoma de Células Renales/diagnóstico por imagen , Carcinoma de Células Renales/cirugía , Estudios Retrospectivos , Recurrencia Local de Neoplasia/cirugía , Nefrectomía , Neoplasias Renales/diagnóstico por imagen , Neoplasias Renales/cirugía , Neoplasias Renales/tratamiento farmacológico , Tomografía Computarizada por Rayos X/métodosRESUMEN
INTRODUCTION: Surgical intervention is important in reducing morbidity and mortality among patients admitted for small bowel obstruction (SBO). Patient-specific variables such as age and comorbidities are risk factors for adverse outcomes after surgery for SBO. However, the effect of weekend admission on outcomes has not been well delineated in the literature. Our aim was to determine whether weekend admission affects mortality and length of stay (LOS) in patients who were admitted for SBO and were managed operatively. MATERIALS AND METHODS: Using the 2006-2012 Nationwide Inpatient Sample (NIS) database, we identified adult patients who were admitted with a primary diagnosis of SBO and had a primary procedure of exploratory laparotomy, lysis of adhesions, or small bowel resection. We performed univariate analysis comparing cases that were admitted on the weekend vs. weekday. We then performed negative binomial regression with LOS as the dependent variable, adjusting for risk variables. RESULTS: 2804 patients were studied, of which 728 (26.0%) were admitted on the weekend. Univariate analysis showed no statistically significant difference in mortality or LOS for patients admitted on a weekday vs. weekend. Multivariate analysis showed that several factors were associated with increased LOS, including third quartile van Walraven score (P < .0001) and large hospital size (P = .0031). Other factors were associated with decreased LOS, including fourth quartile of income (P = .0022) and weekend admission (P = .048). DISCUSSION: There is no significant difference in mortality between patients admitted on weekend vs. weekday for SBO, but patients admitted on weekend are more likely to have a decreased LOS.
Asunto(s)
Mortalidad Hospitalaria , Hospitalización , Obstrucción Intestinal/cirugía , Tiempo de Internación , Anciano , Femenino , Humanos , Masculino , Persona de Mediana Edad , Evaluación de Resultado en la Atención de Salud , Estudios Retrospectivos , Factores de Riesgo , Factores de TiempoRESUMEN
The genetic circuits that allow cancer cells to evade destruction by the host immune system remain poorly understood1-3. Here, to identify a phenotypically robust core set of genes and pathways that enable cancer cells to evade killing mediated by cytotoxic T lymphocytes (CTLs), we performed genome-wide CRISPR screens across a panel of genetically diverse mouse cancer cell lines that were cultured in the presence of CTLs. We identify a core set of 182 genes across these mouse cancer models, the individual perturbation of which increases either the sensitivity or the resistance of cancer cells to CTL-mediated toxicity. Systematic exploration of our dataset using genetic co-similarity reveals the hierarchical and coordinated manner in which genes and pathways act in cancer cells to orchestrate their evasion of CTLs, and shows that discrete functional modules that control the interferon response and tumour necrosis factor (TNF)-induced cytotoxicity are dominant sub-phenotypes. Our data establish a central role for genes that were previously identified as negative regulators of the type-II interferon response (for example, Ptpn2, Socs1 and Adar1) in mediating CTL evasion, and show that the lipid-droplet-related gene Fitm2 is required for maintaining cell fitness after exposure to interferon-γ (IFNγ). In addition, we identify the autophagy pathway as a conserved mediator of the evasion of CTLs by cancer cells, and show that this pathway is required to resist cytotoxicity induced by the cytokines IFNγ and TNF. Through the mapping of cytokine- and CTL-based genetic interactions, together with in vivo CRISPR screens, we show how the pleiotropic effects of autophagy control cancer-cell-intrinsic evasion of killing by CTLs and we highlight the importance of these effects within the tumour microenvironment. Collectively, these data expand our knowledge of the genetic circuits that are involved in the evasion of the immune system by cancer cells, and highlight genetic interactions that contribute to phenotypes associated with escape from killing by CTLs.
Asunto(s)
Genoma/genética , Genómica , Neoplasias/genética , Neoplasias/inmunología , Linfocitos T Citotóxicos/inmunología , Escape del Tumor/genética , Escape del Tumor/inmunología , Animales , Autofagia , Línea Celular Tumoral , Femenino , Genes Relacionados con las Neoplasias/genética , Humanos , Interferón gamma/inmunología , Masculino , Ratones , FN-kappa B/metabolismo , Reproducibilidad de los Resultados , Transducción de SeñalRESUMEN
Patient classification based on clinical and genomic data will further the goal of precision medicine. Interpretability is of particular relevance for models based on genomic data, where sample sizes are relatively small (in the hundreds), increasing overfitting risk netDx is a machine learning method to integrate multi-modal patient data and build a patient classifier. Patient data are converted into networks of patient similarity, which is intuitive to clinicians who also use patient similarity for medical diagnosis. Features passing selection are integrated, and new patients are assigned to the class with the greatest profile similarity. netDx has excellent performance, outperforming most machine-learning methods in binary cancer survival prediction. It handles missing data - a common problem in real-world data - without requiring imputation. netDx also has excellent interpretability, with native support to group genes into pathways for mechanistic insight into predictive features. The netDx Bioconductor package provides multiple workflows for users to build custom patient classifiers. It provides turnkey functions for one-step predictor generation from multi-modal data, including feature selection over multiple train/test data splits. Workflows offer versatility with custom feature design, choice of similarity metric; speed is improved by parallel execution. Built-in functions and examples allow users to compute model performance metrics such as AUROC, AUPR, and accuracy. netDx uses RCy3 to visualize top-scoring pathways and the final integrated patient network in Cytoscape. Advanced users can build more complex predictor designs with functional building blocks used in the default design. Finally, the netDx Bioconductor package provides a novel workflow for pathway-based patient classification from sparse genetic data.
Asunto(s)
Genómica , Programas Informáticos , Humanos , Aprendizaje Automático , Medicina de Precisión , Flujo de TrabajoRESUMEN
Patient classification has widespread biomedical and clinical applications, including diagnosis, prognosis, and treatment response prediction. A clinically useful prediction algorithm should be accurate, generalizable, be able to integrate diverse data types, and handle sparse data. A clinical predictor based on genomic data needs to be interpretable to drive hypothesis-driven research into new treatments. We describe netDx, a novel supervised patient classification framework based on patient similarity networks, which meets these criteria. In a cancer survival benchmark dataset integrating up to six data types in four cancer types, netDx significantly outperforms most other machine-learning approaches across most cancer types. Compared to traditional machine-learning-based patient classifiers, netDx results are more interpretable, visualizing the decision boundary in the context of patient similarity space. When patient similarity is defined by pathway-level gene expression, netDx identifies biological pathways important for outcome prediction, as demonstrated in breast cancer and asthma. netDx can serve as a patient classifier and as a tool for discovery of biological features characteristic of disease. We provide a free software implementation of netDx with automation workflows.
Asunto(s)
Algoritmos , Asma/clasificación , Neoplasias de la Mama/clasificación , Aprendizaje Automático , Programas Informáticos , Asma/diagnóstico , Asma/genética , Benchmarking , Neoplasias de la Mama/diagnóstico , Neoplasias de la Mama/genética , Femenino , Genómica , Humanos , Pronóstico , Análisis de Supervivencia , Flujo de TrabajoRESUMEN
A common complementary strategy in Genome-Wide Association Studies (GWAS) is to perform Gene Set Analysis (GSA), which tests for the association between one phenotype of interest and an entire set of Single Nucleotide Polymorphisms (SNPs) residing in selected genes. While there exist many tools for performing GSA, popular methods often include a number of ad-hoc steps that are difficult to justify statistically, provide complicated interpretations based on permutation inference, and demonstrate poor operating characteristics. Additionally, the lack of gold standard gene set lists can produce misleading results and create difficulties in comparing analyses even across the same phenotype. We introduce the Generalized Berk-Jones (GBJ) statistic for GSA, a permutation-free parametric framework that offers asymptotic power guarantees in certain set-based testing settings. To adjust for confounding introduced by different gene set lists, we further develop a GBJ step-down inference technique that can discriminate between gene sets driven to significance by single genes and those demonstrating group-level effects. We compare GBJ to popular alternatives through simulation and re-analysis of summary statistics from a large breast cancer GWAS, and we show how GBJ can increase power by incorporating information from multiple signals in the same gene. In addition, we illustrate how breast cancer pathway analysis can be confounded by the frequency of FGFR2 in pathway lists. Our approach is further validated on two other datasets of summary statistics generated from GWAS of height and schizophrenia.
Asunto(s)
Estudio de Asociación del Genoma Completo/estadística & datos numéricos , Estatura/genética , Neoplasias de la Mama/genética , Mapeo Cromosómico/estadística & datos numéricos , Biología Computacional/métodos , Simulación por Computador , Bases de Datos Genéticas , Femenino , Redes Reguladoras de Genes , Humanos , Modelos Genéticos , Modelos Estadísticos , Polimorfismo de Nucleótido Simple , Receptor Tipo 2 de Factor de Crecimiento de Fibroblastos/genética , Esquizofrenia/genéticaRESUMEN
In recent years, fluorescent cholangiography using Indocyanine green (ICG) dye has been used to aid identification of structures during robotic cholecystectomy. We sought to compare cholecystectomy with ICG dye versus laparoscopic cholecystectomy at an inner-city academic medical center. Between January 2013 and July 2016, we identified 287 patients of which 191 patients underwent laparoscopic cholecystectomy and 96 patients underwent robotic cholecystectomy with ICG dye. Preoperative risk variables of interest included age, sex, race, body mass index (BMI), and acute cholecystitis. Primary outcome of interest was conversion to open procedures while secondary outcome was length of stay. The two groups were similar in their BMI (31.98 vs. 31.10 kg/m2 for the laparoscopic and robotic, respectively, p = 0.32). The laparoscopic group had a greater mean age compared to the robotic group (47.77 vs. 43.61 years, p = 0.04). There was no significant difference in sex and emergency surgery between the two groups. Fewer open conversions were found in the robotic than the laparoscopic group [2 (2.1%) vs. 17 (8.9%), p = 0.03]. In multiple logistic regression, robotic cholecystectomy with ICG also showed a lower risk of conversion compared to laparoscopic cholecystectomy, but the difference did not reach statistical significance (OR 0.42, 95% CI 0.11-1.65, p = 0.22). ICG fluorescent cholangiography during robotic cholecystectomy may contribute to proper identification of biliary structures and may reduce the rates of open conversion. The preliminary results of fewer open conversions are promising. Further studies with a large randomized prospective controlled study should be taken for further evaluation.
Asunto(s)
Colangiografía/métodos , Colecistectomía/métodos , Angiografía con Fluoresceína/métodos , Procedimientos Quirúrgicos Robotizados/métodos , Adulto , Anciano , Femenino , Colorantes Fluorescentes/uso terapéutico , Enfermedades de la Vesícula Biliar/cirugía , Humanos , Masculino , Persona de Mediana Edad , Estudios RetrospectivosRESUMEN
Most common breast cancer susceptibility variants have been identified through genome-wide association studies (GWAS) of predominantly estrogen receptor (ER)-positive disease. We conducted a GWAS using 21,468 ER-negative cases and 100,594 controls combined with 18,908 BRCA1 mutation carriers (9,414 with breast cancer), all of European origin. We identified independent associations at P < 5 × 10-8 with ten variants at nine new loci. At P < 0.05, we replicated associations with 10 of 11 variants previously reported in ER-negative disease or BRCA1 mutation carrier GWAS and observed consistent associations with ER-negative disease for 105 susceptibility variants identified by other studies. These 125 variants explain approximately 16% of the familial risk of this breast cancer subtype. There was high genetic correlation (0.72) between risk of ER-negative breast cancer and breast cancer risk for BRCA1 mutation carriers. These findings may lead to improved risk prediction and inform further fine-mapping and functional work to better understand the biological basis of ER-negative breast cancer.
Asunto(s)
Proteína BRCA1/genética , Neoplasias de la Mama/genética , Predisposición Genética a la Enfermedad/genética , Mutación , Polimorfismo de Nucleótido Simple , Neoplasias de la Mama/etnología , Neoplasias de la Mama/metabolismo , Femenino , Predisposición Genética a la Enfermedad/etnología , Estudio de Asociación del Genoma Completo/métodos , Heterocigoto , Humanos , Receptores de Estrógenos/metabolismo , Factores de Riesgo , Población Blanca/genéticaRESUMEN
Breast cancer risk is influenced by rare coding variants in susceptibility genes, such as BRCA1, and many common, mostly non-coding variants. However, much of the genetic contribution to breast cancer risk remains unknown. Here we report the results of a genome-wide association study of breast cancer in 122,977 cases and 105,974 controls of European ancestry and 14,068 cases and 13,104 controls of East Asian ancestry. We identified 65 new loci that are associated with overall breast cancer risk at P < 5 × 10-8. The majority of credible risk single-nucleotide polymorphisms in these loci fall in distal regulatory elements, and by integrating in silico data to predict target genes in breast cells at each locus, we demonstrate a strong overlap between candidate target genes and somatic driver genes in breast tumours. We also find that heritability of breast cancer due to all single-nucleotide polymorphisms in regulatory features was 2-5-fold enriched relative to the genome-wide average, with strong enrichment for particular transcription factor binding sites. These results provide further insight into genetic susceptibility to breast cancer and will improve the use of genetic risk scores for individualized screening and prevention.
Asunto(s)
Neoplasias de la Mama/genética , Sitios Genéticos , Predisposición Genética a la Enfermedad/genética , Estudio de Asociación del Genoma Completo , Asia/etnología , Pueblo Asiatico/genética , Sitios de Unión/genética , Neoplasias de la Mama/diagnóstico , Simulación por Computador , Europa (Continente)/etnología , Femenino , Humanos , Herencia Multifactorial/genética , Polimorfismo de Nucleótido Simple/genética , Secuencias Reguladoras de Ácidos Nucleicos , Medición de Riesgo , Factores de Transcripción/metabolismo , Población Blanca/genéticaRESUMEN
Aim. To study serum triglyceride level as a predictor of complications and outcomes in acute pancreatitis. Methods. In this retrospective observational study, 582 patients admitted with acute pancreatitis, who had serum triglyceride levels measured within the first 24 hours, were divided into two groups. The study group consisted of patients with a triglyceride level ≥2.26 mmol/L (group 2) and the control group consisted of triglyceride level of <2.26 mmol/L (group 1). We collected data for baseline demographics, laboratory values, incidence of complications (local and systemic), admission to the intensive care unit (ICU), ICU length of stay, length of total hospital stay, and death in the two groups. Results. A triglyceride level of ≥2.26 mmol/L was found to be an independent predictor of developing altered mental status (p: 0.004), pancreatic necrosis (p: 0.001), acute respiratory distress syndrome (p: 0001), systemic Inflammatory response syndrome (p: 0.001), acute kidney injury (p: 0.001), hospital length of stay (LOS) (p: 0.002), admission to intensive care unit (ICU) (p: 0.002), and ICU LOS (p: 0.003). Conclusion. A triglyceride level of ≥2.26 mmol/L on admission in acute pancreatitis is an independent predictor of developing local and systemic complications, hospital LOS, admission to ICU, and ICU LOS.
Asunto(s)
Hospitalización , Unidades de Cuidados Intensivos/estadística & datos numéricos , Pancreatitis/sangre , Triglicéridos/sangre , Enfermedad Aguda , Adulto , Femenino , Humanos , Tiempo de Internación , Masculino , Persona de Mediana Edad , Pancreatitis/complicaciones , Valor Predictivo de las Pruebas , Pronóstico , Estudios Retrospectivos , Factores de RiesgoRESUMEN
BACKGROUND: PDZ domains are structural protein domains that recognize simple linear amino acid motifs, often at protein C-termini, and mediate protein-protein interactions (PPIs) in important biological processes, such as ion channel regulation, cell polarity and neural development. PDZ domain-peptide interaction predictors have been developed based on domain and peptide sequence information. Since domain structure is known to influence binding specificity, we hypothesized that structural information could be used to predict new interactions compared to sequence-based predictors. RESULTS: We developed a novel computational predictor of PDZ domain and C-terminal peptide interactions using a support vector machine trained with PDZ domain structure and peptide sequence information. Performance was estimated using extensive cross validation testing. We used the structure-based predictor to scan the human proteome for ligands of 218 PDZ domains and show that the predictions correspond to known PDZ domain-peptide interactions and PPIs in curated databases. The structure-based predictor is complementary to the sequence-based predictor, finding unique known and novel PPIs, and is less dependent on training-testing domain sequence similarity. We used a functional enrichment analysis of our hits to create a predicted map of PDZ domain biology. This map highlights PDZ domain involvement in diverse biological processes, some only found by the structure-based predictor. Based on this analysis, we predict novel PDZ domain involvement in xenobiotic metabolism and suggest new interactions for other processes including wound healing and Wnt signalling. CONCLUSIONS: We built a structure-based predictor of PDZ domain-peptide interactions, which can be used to scan C-terminal proteomes for PDZ interactions. We also show that the structure-based predictor finds many known PDZ mediated PPIs in human that were not found by our previous sequence-based predictor and is less dependent on training-testing domain sequence similarity. Using both predictors, we defined a functional map of human PDZ domain biology and predict novel PDZ domain function. Users may access our structure-based and previous sequence-based predictors at http://webservice.baderlab.org/domains/POW.
Asunto(s)
Dominios PDZ , Mapeo de Interacción de Proteínas/métodos , Humanos , Ligandos , Modelos Moleculares , Péptidos/química , Péptidos/metabolismo , Mapas de Interacción de Proteínas , Proteoma/química , Proteoma/metabolismo , Análisis de Secuencia de Proteína , Máquina de Vectores de SoporteRESUMEN
Protein-protein interactions (PPIs), involved in many biological processes such as cellular signaling, are ultimately encoded in the genome. Solving the problem of predicting protein interactions from the genome sequence will lead to increased understanding of complex networks, evolution and human disease. We can learn the relationship between genomes and networks by focusing on an easily approachable subset of high-resolution protein interactions that are mediated by peptide recognition modules (PRMs) such as PDZ, WW and SH3 domains. This review focuses on computational prediction and analysis of PRM-mediated networks and discusses sequence- and structure-based interaction predictors, techniques and datasets for identifying physiologically relevant PPIs, and interpreting high-resolution interaction networks in the context of evolution and human disease.
Asunto(s)
Genoma , Mutación , Mapeo de Interacción de Proteínas/métodos , Secuencia de Aminoácidos , Animales , Inteligencia Artificial , Biología Computacional/métodos , ADN/química , Humanos , Conformación Molecular , Datos de Secuencia Molecular , Péptidos/química , Unión Proteica , Dominios y Motivos de Interacción de Proteínas , Estructura Terciaria de Proteína , Proteínas/química , Dominios Homologos srcRESUMEN
BACKGROUND: PDZ domains mediate protein-protein interactions involved in important biological processes through the recognition of short linear motifs in their target proteins. Two recent independent studies have used protein microarray or phage display technology to detect PDZ domain interactions with peptide ligands on a large scale. Several computational predictors of PDZ domain interactions have been developed, however they are trained using only protein microarray data and focus on limited subsets of PDZ domains. An accurate predictor of genomic PDZ domain interactions would allow the proteomes of organisms to be scanned for potential binders. Such an application would require an accurate and precise predictor to avoid generating too many false positive hits given the large amount of possible interactors in a given proteome. Once validated these predictions will help to increase the coverage of current PDZ domain interaction networks and further our understanding of the roles that PDZ domains play in a variety of biological processes. RESULTS: We developed a PDZ domain interaction predictor using a support vector machine (SVM) trained with both protein microarray and phage display data. In order to use the phage display data for training, which only contains positive interactions, we developed a method to generate artificial negative interactions. Using cross-validation and a series of independent tests, we showed that our SVM successfully predicts interactions in different organisms. We then used the SVM to scan the proteomes of human, worm and fly to predict binders for several PDZ domains. Predictions were validated using known genomic interactions and published protein microarray experiments. Based on our results, new protein interactions potentially associated with Usher and Bardet-Biedl syndromes were predicted. A comparison of performance measures (F1 measure and FPR) for the SVM and published predictors demonstrated our SVM's improved accuracy and precision at proteome scanning. CONCLUSIONS: We built an SVM using mouse and human experimental training data to predict PDZ domain interactions. We showed that it correctly predicts known interactions from proteomes of different organisms and is more accurate and precise at proteome scanning compared with published state-of-the-art predictors.
Asunto(s)
Inteligencia Artificial , Dominios PDZ , Proteoma/química , Animales , Sitios de Unión , Humanos , Ratones , Análisis por Matrices de Proteínas/métodos , Mapeo de Interacción de Proteínas/métodos , Proteínas/química , Proteínas/metabolismo , Proteoma/metabolismoRESUMEN
Multicellular organisms rely on complex, fine-tuned protein networks to respond to environmental changes. We used in vitro evolution to explore the role of domain mutation and expansion in the evolution of network complexity. Using random mutagenesis to facilitate family expansion, we asked how versatile and robust the binding site must be to produce the rich functional diversity of the natural PDZ domain family. From a combinatorial protein library, we analyzed several hundred structured domain variants and found that one-quarter were functional for carboxyl-terminal ligand recognition and that our variant repertoire was as specific and diverse as the natural family. Our results show that ligand binding is hardwired in the PDZ fold and suggest that this flexibility may facilitate the rapid evolution of complex protein interaction networks.