Mechanical feedback through E-cadherin promotes direction sensing during collective cell migration.

E-cadherin is a major homophilic cell-cell adhesion molecule that inhibits motility of individual cells on matrix. However, its contribution to migration of cells through cell-rich tissues is less clear. We developed an in vivo sensor of mechanical tension across E-cadherin molecules, which we combined with cell-type-specific RNAi, photoactivatable Rac, and morphodynamic profiling, to interrogate how E-cadherin contributes to collective migration of cells between other cells. Using the Drosophila ovary as a model, we found that adhesion between border cells and their substrate, the nurse cells, functions in a positive feedback loop with Rac and actin assembly to stabilize forward-directed protrusion and directionally persistent movement. Adhesion between individual border cells communicates direction from the lead cell to the followers. Adhesion between motile cells and polar cells holds the cluster together and polarizes each individual cell. Thus, E-cadherin is an integral component of the guidance mechanisms that orchestrate collective chemotaxis in vivo.

Identification of stem cell populations in sweat glands and ducts reveals roles in homeostasis and wound repair.

Sweat glands are abundant in the body and essential for thermoregulation. Like mammary glands, they originate from epidermal progenitors. However, they display few signs of cellular turnover, and whether they have stem cells and tissue-regenerative capacity remains largely unexplored. Using lineage tracing, we here identify in sweat ducts multipotent progenitors that transition to unipotency after developing the sweat gland. In characterizing four adult stem cell populations of glandular skin, we show that they display distinct regenerative capabilities and remain unipotent when healing epidermal, myoepithelial-specific, and lumenal-specific injuries. We devise purification schemes and isolate and transcriptionally profile progenitors. Exploiting molecular differences between sweat and mammary glands, we show that only some progenitors regain multipotency to produce de novo ductal and glandular structures, but that these can retain their identity even within certain foreign microenvironments. Our findings provide insight into glandular stem cells and a framework for the further study of sweat gland biology.

LIN28 cooperates with WNT signaling to drive invasive intestinal and colorectal adenocarcinoma in mice and humans.

Colorectal cancer (CRC) remains a major contributor to cancer-related mortality. LIN28A and LIN28B are highly related RNA-binding protein paralogs that regulate biogenesis of let-7 microRNAs and influence development, metabolism, tissue regeneration, and oncogenesis. Here we demonstrate that overexpression of either LIN28 paralog cooperates with the Wnt pathway to promote invasive intestinal adenocarcinoma in murine models. When LIN28 alone is induced genetically, half of the resulting tumors harbor Ctnnb1 (ß-catenin) mutation. When overexpressed in Apc(Min/+) mice, LIN28 accelerates tumor formation and enhances proliferation and invasiveness. In conditional genetic models, enforced expression of a LIN28-resistant form of the let-7 microRNA reduces LIN28-induced tumor burden, while silencing of LIN28 expression reduces tumor volume and increases tumor differentiation, indicating that LIN28 contributes to tumor maintenance. We detected aberrant expression of LIN28A and/or LIN28B in 38% of a large series of human CRC samples (n = 595), where LIN28 expression levels were associated with invasive tumor growth. Our late-stage CRC murine models and analysis of primary human tumors demonstrate prominent roles for both LIN28 paralogs in promoting CRC growth and progression and implicate the LIN28/let-7 pathway as a therapeutic target.

Whole-genome sequencing of Atacama skeleton shows novel mutations linked with dysplasia.

Over a decade ago, the Atacama humanoid skeleton (Ata) was discovered in the Atacama region of Chile. The Ata specimen carried a strange phenotype-6-in stature, fewer than expected ribs, elongated cranium, and accelerated bone age-leading to speculation that this was a preserved nonhuman primate, human fetus harboring genetic mutations, or even an extraterrestrial. We previously reported that it was human by DNA analysis with an estimated bone age of about 6-8 yr at the time of demise. To determine the possible genetic drivers of the observed morphology, DNA from the specimen was subjected to whole-genome sequencing using the Illumina HiSeq platform with an average 11.5× coverage of 101-bp, paired-end reads. In total, 3,356,569 single nucleotide variations (SNVs) were found as compared to the human reference genome, 518,365 insertions and deletions (indels), and 1047 structural variations (SVs) were detected. Here, we present the detailed whole-genome analysis showing that Ata is a female of human origin, likely of Chilean descent, and its genome harbors mutations in genes (COL1A1, COL2A1, KMT2D, FLNB, ATR, TRIP11, PCNT) previously linked with diseases of small stature, rib anomalies, cranial malformations, premature joint fusion, and osteochondrodysplasia (also known as skeletal dysplasia). Together, these findings provide a molecular characterization of Ata's peculiar phenotype, which likely results from multiple known and novel putative gene mutations affecting bone development and ossification.

The TCF-1 and LEF-1 transcription factors have cooperative and opposing roles in T cell development and malignancy.

The TCF-1 and LEF-1 transcription factors are known to play critical roles in normal thymocyte development. Unexpectedly, we found that TCF-1-deficient (Tcf7(-/-)) mice developed aggressive T cell malignancy, resembling human T cell acute lymphoblastic leukemia (T-ALL). LEF-1 was aberrantly upregulated in premalignant Tcf7(-/-) early thymocytes and lymphoma cells. We further demonstrated that TCF-1 directly repressed LEF-1 expression in early thymocytes and that conditional inactivation of Lef1 greatly delayed or prevented T cell malignancy in Tcf7(-/-) mice. In human T-ALLs, an early thymic progenitor (ETP) subtype was associated with diminished TCF7 expression, and two of the ETP-ALL cases harbored TCF7 gene deletions. We also showed that TCF-1 and LEF-1 were dispensable for T cell lineage commitment but instead were required for early thymocytes to mature beyond the CD4(-)CD8(-) stage. TCF-1 thus has dual roles, i.e., acting cooperatively with LEF-1 to promote thymocyte maturation while restraining LEF-1 expression to prevent malignant transformation of developing thymocytes.

Loss of oncogenic Notch1 with resistance to a PI3K inhibitor in T-cell leukaemia.

Mutations that deregulate Notch1 and Ras/phosphoinositide 3 kinase (PI3K)/Akt signalling are prevalent in T-cell acute lymphoblastic leukaemia (T-ALL), and often coexist. Here we show that the PI3K inhibitor GDC-0941 is active against primary T-ALLs from wild-type and Kras(G12D) mice, and addition of the MEK inhibitor PD0325901 increases its efficacy. Mice invariably relapsed after treatment with drug-resistant clones, most of which unexpectedly had reduced levels of activated Notch1 protein, downregulated many Notch1 target genes, and exhibited cross-resistance to γ-secretase inhibitors. Multiple resistant primary T-ALLs that emerged in vivo did not contain somatic Notch1 mutations present in the parental leukaemia. Importantly, resistant clones upregulated PI3K signalling. Consistent with these data, inhibiting Notch1 activated the PI3K pathway, providing a likely mechanism for selection against oncogenic Notch1 signalling. These studies validate PI3K as a therapeutic target in T-ALL and raise the unexpected possibility that dual inhibition of PI3K and Notch1 signalling could promote drug resistance in T-ALL.

Oncogenic Nras has bimodal effects on stem cells that sustainably increase competitiveness.

'Pre-leukaemic' mutations are thought to promote clonal expansion of haematopoietic stem cells (HSCs) by increasing self-renewal and competitiveness; however, mutations that increase HSC proliferation tend to reduce competitiveness and self-renewal potential, raising the question of how a mutant HSC can sustainably outcompete wild-type HSCs. Activating mutations in NRAS are prevalent in human myeloproliferative neoplasms and leukaemia. Here we show that a single allele of oncogenic Nras(G12D) increases HSC proliferation but also increases reconstituting and self-renewal potential upon serial transplantation in irradiated mice, all prior to leukaemia initiation. Nras(G12D) also confers long-term self-renewal potential to multipotent progenitors. To explore the mechanism by which Nras(G12D) promotes HSC proliferation and self-renewal, we assessed cell-cycle kinetics using H2B-GFP label retention and 5-bromodeoxyuridine (BrdU) incorporation. Nras(G12D) had a bimodal effect on HSCs, increasing the frequency with which some HSCs divide and reducing the frequency with which others divide. This mirrored bimodal effects on reconstituting potential, as rarely dividing Nras(G12D) HSCs outcompeted wild-type HSCs, whereas frequently dividing Nras(G12D) HSCs did not. Nras(G12D) caused these effects by promoting STAT5 signalling, inducing different transcriptional responses in different subsets of HSCs. One signal can therefore increase HSC proliferation, competitiveness and self-renewal through bimodal effects on HSC gene expression, cycling and reconstituting potential.

The genetic basis of early T-cell precursor acute lymphoblastic leukaemia.

Early T-cell precursor acute lymphoblastic leukaemia (ETP ALL) is an aggressive malignancy of unknown genetic basis. We performed whole-genome sequencing of 12 ETP ALL cases and assessed the frequency of the identified somatic mutations in 94 T-cell acute lymphoblastic leukaemia cases. ETP ALL was characterized by activating mutations in genes regulating cytokine receptor and RAS signalling (67% of cases; NRAS, KRAS, FLT3, IL7R, JAK3, JAK1, SH2B3 and BRAF), inactivating lesions disrupting haematopoietic development (58%; GATA3, ETV6, RUNX1, IKZF1 and EP300) and histone-modifying genes (48%; EZH2, EED, SUZ12, SETD2 and EP300). We also identified new targets of recurrent mutation including DNM2, ECT2L and RELN. The mutational spectrum is similar to myeloid tumours, and moreover, the global transcriptional profile of ETP ALL was similar to that of normal and myeloid leukaemia haematopoietic stem cells. These findings suggest that addition of myeloid-directed therapies might improve the poor outcome of ETP ALL.

Targetable kinase-activating lesions in Ph-like acute lymphoblastic leukemia.

BACKGROUND: Philadelphia chromosome-like acute lymphoblastic leukemia (Ph-like ALL) is characterized by a gene-expression profile similar to that of BCR-ABL1-positive ALL, alterations of lymphoid transcription factor genes, and a poor outcome. The frequency and spectrum of genetic alterations in Ph-like ALL and its responsiveness to tyrosine kinase inhibition are undefined, especially in adolescents and adults. METHODS: We performed genomic profiling of 1725 patients with precursor B-cell ALL and detailed genomic analysis of 154 patients with Ph-like ALL. We examined the functional effects of fusion proteins and the efficacy of tyrosine kinase inhibitors in mouse pre-B cells and xenografts of human Ph-like ALL. RESULTS: Ph-like ALL increased in frequency from 10% among children with standard-risk ALL to 27% among young adults with ALL and was associated with a poor outcome. Kinase-activating alterations were identified in 91% of patients with Ph-like ALL; rearrangements involving ABL1, ABL2, CRLF2, CSF1R, EPOR, JAK2, NTRK3, PDGFRB, PTK2B, TSLP, or TYK2 and sequence mutations involving FLT3, IL7R, or SH2B3 were most common. Expression of ABL1, ABL2, CSF1R, JAK2, and PDGFRB fusions resulted in cytokine-independent proliferation and activation of phosphorylated STAT5. Cell lines and human leukemic cells expressing ABL1, ABL2, CSF1R, and PDGFRB fusions were sensitive in vitro to dasatinib, EPOR and JAK2 rearrangements were sensitive to ruxolitinib, and the ETV6-NTRK3 fusion was sensitive to crizotinib. CONCLUSIONS: Ph-like ALL was found to be characterized by a range of genomic alterations that activate a limited number of signaling pathways, all of which may be amenable to inhibition with approved tyrosine kinase inhibitors. Trials identifying Ph-like ALL are needed to assess whether adding tyrosine kinase inhibitors to current therapy will improve the survival of patients with this type of leukemia. (Funded by the American Lebanese Syrian Associated Charities and others.).

Defective K-Ras oncoproteins overcome impaired effector activation to initiate leukemia in vivo.

Reversing the aberrant biochemical output of oncogenic Ras proteins is one of the great challenges in cancer therapeutics; however, it is uncertain which Ras effectors are required for tumor initiation and maintenance. To address this question, we expressed oncogenic K-Ras(D12) proteins with "second site" amino acid substitutions that impair PI3 kinase/Akt or Raf/MEK/ERK activation in bone marrow cells and transplanted them into recipient mice. In spite of attenuated signaling properties, defective K-Ras oncoproteins initiated aggressive clonal T-lineage acute lymphoblastic leukemia (T-ALL). Murine T-ALLs expressing second site mutant proteins restored full oncogenic Ras activity through diverse mechanisms, which included acquiring novel somatic third site Kras(D12) mutations and silencing PTEN. T-ALL cell lines lacking PTEN had elevated levels of phosphorylated Akt, a gene expression pattern similar to human early T-cell precursor ALL, and were resistant to the potent and selective MEK inhibitor PD0325901. Our data, which demonstrate strong selective pressure to overcome the defective activation of PI3 kinase/Akt and Raf/MEK/ERK, implicate both Ras effector pathways as drivers of aberrant growth in T-ALL and further suggest that leukemia cells will deploy multiple mechanisms to develop resistance to targeted inhibitors in vivo.

Knockdown of Hnrnpa0, a del(5q) gene, alters myeloid cell fate in murine cells through regulation of AU-rich transcripts.

The control of mRNA stability plays a central role in orchestrating gene-regulatory networks in hematopoietic cell growth, differentiation and tumorigenesis. HNRNPA0, which encodes an RNA-binding protein shown to regulate transcript stability via binding to the AU-rich elements of mRNAs, is located within the commonly deleted segment of 5q31.2 in myeloid neoplasms with a del(5q), and is expressed at haploinsufficient levels in these patients. We show that HNRNPA0 is normally highly expressed in hematopoietic stem cells and exhibits dynamic changes in expression during the course of differentiation. To model HNRNPA0 haploinsufficiency, we used RNAi interference in primary murine cells and an experimental cell system, and found that reduced Hnrnpa0 expression leads to a shift from monocytic towards granulocytic differentiation. Microarray-based global expression profiling revealed that Hnrnpa0 knockdown disproportionally impacts AU-rich containing transcripts and alters expression of myeloid specification genes. In therapy-related myeloid neoplasms with a del(5q), AU-rich containing mRNAs are enriched in transcripts that encode proteins associated with increased growth and proliferation. Our findings implicate haploinsufficiency of HNRNPA0 as one of the key initiating mutations in the pathogenesis of myeloid neoplasms with a del(5q), and suggest that therapies that target AU-rich elements warrant consideration in efforts to develop new mechanism-based treatment strategies.

The BCL11B tumor suppressor is mutated across the major molecular subtypes of T-cell acute lymphoblastic leukemia.

The BCL11B transcription factor is required for normal T-cell development, and has recently been implicated in the pathogenesis of T-cell acute lymphoblastic leukemia (T-ALL) induced by TLX overexpression or Atm deficiency. To comprehensively assess the contribution of BCL11B inactivation to human T-ALL, we performed DNA copy number and sequencing analyses of T-ALL diagnostic specimens, revealing monoallelic BCL11B deletions or missense mutations in 9% (n = 10 of 117) of cases. Structural homology modeling revealed that several of the BCL11B mutations disrupted the structure of zinc finger domains required for this transcription factor to bind DNA. BCL11B haploinsufficiency occurred across each of the major molecular subtypes of T-ALL, including early T-cell precursor, HOXA-positive, LEF1-inactivated, and TAL1-positive subtypes, which have differentiation arrest at diverse stages of thymocyte development. Our findings provide compelling evidence that BCL11B is a haploinsufficient tumor suppressor that collaborates with all major T-ALL oncogenic lesions in human thymocyte transformation.

The Feasibility and Performance of Total Hip Replacement Prediction Deep Learning Algorithm with Real World Data.

(1) Background: Hip degenerative disorder is a common geriatric disease is the main causes to lead to total hip replacement (THR). The surgical timing of THR is crucial for post-operative recovery. Deep learning (DL) algorithms can be used to detect anomalies in medical images and predict the need for THR. The real world data (RWD) were used to validate the artificial intelligence and DL algorithm in medicine but there was no previous study to prove its function in THR prediction. (2) Methods: We designed a sequential two-stage hip replacement prediction deep learning algorithm to identify the possibility of THR in three months of hip joints by plain pelvic radiography (PXR). We also collected RWD to validate the performance of this algorithm. (3) Results: The RWD totally included 3766 PXRs from 2018 to 2019. The overall accuracy of the algorithm was 0.9633; sensitivity was 0.9450; specificity was 1.000 and the precision was 1.000. The negative predictive value was 0.9009, the false negative rate was 0.0550, and the F1 score was 0.9717. The area under curve was 0.972 with 95% confidence interval from 0.953 to 0.987. (4) Conclusions: In summary, this DL algorithm can provide an accurate and reliable method for detecting hip degeneration and predicting the need for further THR. RWD offered an alternative support of the algorithm and validated its function to save time and cost.

Predicting the Risk of Total Hip Replacement by Using A Deep Learning Algorithm on Plain Pelvic Radiographs: Diagnostic Study.

BACKGROUND: Total hip replacement (THR) is considered the gold standard of treatment for refractory degenerative hip disorders. Identifying patients who should receive THR in the short term is important. Some conservative treatments, such as intra-articular injection administered a few months before THR, may result in higher odds of arthroplasty infection. Delayed THR after functional deterioration may result in poorer outcomes and longer waiting times for those who have been flagged as needing THR. Deep learning (DL) in medical imaging applications has recently obtained significant breakthroughs. However, the use of DL in practical wayfinding, such as short-term THR prediction, is still lacking. OBJECTIVE: In this study, we will propose a DL-based assistant system for patients with pelvic radiographs to identify the need for THR within 3 months. METHODS: We developed a convolutional neural network-based DL algorithm to analyze pelvic radiographs, predict the hip region of interest (ROI), and determine whether or not THR is required. The data set was collected from August 2008 to December 2017. The images included 3013 surgical hip ROIs that had undergone THR and 1630 nonsurgical hip ROIs. The images were split, using split-sample validation, into training (n=3903, 80%), validation (n=476, 10%), and testing (n=475, 10%) sets to evaluate the algorithm performance. RESULTS: The algorithm, called SurgHipNet, yielded an area under the receiver operating characteristic curve of 0.994 (95% CI 0.990-0.998). The accuracy, sensitivity, specificity, and F1-score of the model were 0.977, 0.920, 0932, and 0.944, respectively. CONCLUSIONS: The proposed approach has demonstrated that SurgHipNet shows the ability and potential to provide efficient support in clinical decision-making; it can assist physicians in promptly determining the optimal timing for THR.

Automatic Hip Detection in Anteroposterior Pelvic Radiographs-A Labelless Practical Framework.

Automated detection of the region of interest (ROI) is a critical step in the two-step classification system in several medical image applications. However, key information such as model parameter selection, image annotation rules, and ROI confidence score are essential but usually not reported. In this study, we proposed a practical framework of ROI detection by analyzing hip joints seen on 7399 anteroposterior pelvic radiographs (PXR) from three diverse sources. We presented a deep learning-based ROI detection framework utilizing a single-shot multi-box detector with a customized head structure based on the characteristics of the obtained datasets. Our method achieved average intersection over union (IoU) = 0.8115, average confidence = 0.9812, and average precision with threshold IoU = 0.5 (AP50) = 0.9901 in the independent testing set, suggesting that the detected hip regions appropriately covered the main features of the hip joints. The proposed approach featured flexible loose-fitting labeling, customized model design, and heterogeneous data testing. We demonstrated the feasibility of training a robust hip region detector for PXRs. This practical framework has a promising potential for a wide range of medical image applications.

Automated image analysis of protein localization in budding yeast.

MOTIVATION: The yeast Saccharomyces cerevisiae is the first eukaryotic organism to have its genome completely sequenced. Since then, several large-scale analyses of the yeast genome have provided extensive functional annotations of individual genes and proteins. One fundamental property of a protein is its subcellular localization, which provides critical information about how this protein works in a cell. An important project therefore was the creation of the yeast GFP fusion localization database by the University of California, San Francisco, USA (UCSF). This database provides localization data for 75% of the proteins believed to be encoded by the yeast genome. These proteins were classified into 22 distinct subcellular location categories by visual examination. Based on our past success at building automated systems to classify subcellular location patterns in mammalian cells, we sought to create a similar system for yeast. RESULTS: We developed computational methods to automatically analyze the images created by the UCSF yeast GFP fusion localization project. The system was trained to recognize the same location categories that were used in that study. We applied the system to 2640 images, and the system gave the same label as the previous assignments to 2139 images (81%). When only the highest confidence assignments were considered, 94.7% agreement was observed. Visual examination of the proteins for which the two approaches disagree suggests that at least some of the automated assignments may be more accurate. The automated method provides an objective, quantitative and repeatable assignment of protein locations that can be applied to new collections of yeast images (e.g. for different strains or the same strain under different conditions). It is also important to note that this performance could be achieved without requiring colocalization with any marker proteins. AVAILABILITY: The original images analyzed in this article are available at http://yeastgfp.ucsf.edu, and source code and results are available at http://murphylab.web.cmu.edu/software.

oGNM: online computation of structural dynamics using the Gaussian Network Model.

An assessment of the equilibrium dynamics of biomolecular systems, and in particular their most cooperative fluctuations accessible under native state conditions, is a first step towards understanding molecular mechanisms relevant to biological function. We present a web-based system, oGNM that enables users to calculate online the shape and dispersion of normal modes of motion for proteins, oligonucleotides and their complexes, or associated biological units, using the Gaussian Network Model (GNM). Computations with the new engine are 5-6 orders of magnitude faster than those using conventional normal mode analyses. Two cases studies illustrate the utility of oGNM. The first shows that the thermal fluctuations predicted for 1250 non-homologous proteins correlate well with X-ray crystallographic data over a broad range [7.3-15 A] of inter-residue interaction cutoff distances and the correlations improve with increasing observation temperatures. The second study, focused on 64 oligonucleotides and oligonucleotide-protein complexes, shows that good agreement with experiments is achieved by representing each nucleotide by three GNM nodes (as opposed to one-node-per-residue in proteins) along with uniform interaction ranges for all components of the complexes. These results open the way to a rapid assessment of the dynamics of DNA/RNA-containing complexes. The server can be accessed at http://ignm.ccbb.pitt.edu/GNM_Online_Calculation.htm.

A graphical model approach to automated classification of protein subcellular location patterns in multi-cell images.

BACKGROUND: Knowledge of the subcellular location of a protein is critical to understanding how that protein works in a cell. This location is frequently determined by the interpretation of fluorescence microscope images. In recent years, automated systems have been developed for consistent and objective interpretation of such images so that the protein pattern in a single cell can be assigned to a known location category. While these systems perform with nearly perfect accuracy for single cell images of all major subcellular structures, their ability to distinguish subpatterns of an organelle (such as two Golgi proteins) is not perfect. Our goal in the work described here was to improve the ability of an automated system to decide which of two similar patterns is present in a field of cells by considering more than one cell at a time. Since cells displaying the same location pattern are often clustered together, considering multiple cells may be expected to improve discrimination between similar patterns. RESULTS: We describe how to take advantage of information on experimental conditions to construct a graphical representation for multiple cells in a field. Assuming that a field is composed of a small number of classes, the classification accuracy can be improved by allowing the computed probability of each pattern for each cell to be influenced by the probabilities of its neighboring cells in the model. We describe a novel way to allow this influence to occur, in which we adjust the prior probabilities of each class to reflect the patterns that are present. When this graphical model approach is used on synthetic multi-cell images in which the true class of each cell is known, we observe that the ability to distinguish similar classes is improved without suffering any degradation in ability to distinguish dissimilar classes. The computational complexity of the method is sufficiently low that improved assignments of classes can be obtained for fields of twelve cells in under 0.04 second on a 1600 megahertz processor. CONCLUSION: We demonstrate that graphical models can be used to improve the accuracy of classification of subcellular patterns in multi-cell fluorescence microscope images. We also describe a novel algorithm for inferring classes from a graphical model. The performance and speed suggest that the method will be particularly valuable for analysis of images from high-throughput microscopy. We also anticipate that it will be useful for analyzing the mixtures of cell types typically present in images of tissues. Lastly, we anticipate that the method can be generalized to other problems.

Mining frequent patterns in protein structures: a study of protease families.

MOTIVATION: Analysis of protein sequence and structure databases usually reveal frequent patterns (FP) associated with biological function. Data mining techniques generally consider the physicochemical and structural properties of amino acids and their microenvironment in the folded structures. Dynamics is not usually considered, although proteins are not static, and their function relates to conformational mobility in many cases. RESULTS: This work describes a novel unsupervised learning approach to discover FPs in the protein families, based on biochemical, geometric and dynamic features. Without any prior knowledge of functional motifs, the method discovers the FPs for each type of amino acid and identifies the conserved residues in three protease subfamilies; chymotrypsin and subtilisin subfamilies of serine proteases and papain subfamily of cysteine proteases. The catalytic triad residues are distinguished by their strong spatial coupling (high interconnectivity) to other conserved residues. Although the spatial arrangements of the catalytic residues in the two subfamilies of serine proteases are similar, their FPs are found to be quite different. The present approach appears to be a promising tool for detecting functional patterns in rapidly growing structure databases and providing insights in to the relationship among protein structure, dynamics and function. AVAILABILITY: Available upon request from the authors.