Evaluating a novel patient pathway to manage symptomatic breast referrals (the blue flag clinic): a longitudinal observational study

INTRODUCTION

A novel referral pathway for exhibited breast symptom (EBS) referrals to manage increasing referrals of urgent suspected cancer (USC) was implemented in our trust. We report on the safety and effect on compliance with the 2-week-wait rule (2WW).

METHODS

A single-centre longitudinal observational study included all patients referred to a UK breast unit during 13 May 2019 to 27 March 2020 (period 1) and 8 February 2021 to 31 January 2022 (period 2). USC referrals were assessed in a one-stop clinic (red flag clinic [RFC]); EBS referrals were assessed in a new clinic in which clinical evaluation was performed and imaging occurred subsequently (blue flag clinic [BFC]). Patients were followed up to determine the symptomatic interval cancer rate.

RESULTS

There were 9,695 referrals; 1,655 referrals (17%) were assessed in the BFC after 63 exclusions. Some 95.9% of patients had a benign clinical examination (P1/P2), 80.1% had imaging (mammogram or ultrasound) and 4% had a tissue biopsy. In total, 16/1,655 (0.97%) BFC patients and 510/7,977 (8.2%) RFC patients were diagnosed with breast cancer (breast cancer detection rate). Some 1,631 patients (with 1,639 referrals) were discharged and followed up for a median of 17 months (interquartile range 12-32) with one subsequent cancer diagnosis (symptomatic interval cancer rate, 0.06%). Implementation of the BFC pathway increased 3-month average trust performance of USC referrals with 2WW standard from 8.5% to 98.7% (period 1) and from 30% to 66% (period 2).

CONCLUSIONS

The BFC pathway for EBS patients is safe and implementation led to improvement against the 2WW target for USC referrals, ensuring resources are prioritised to patients with the highest likelihood of breast cancer.

The vertebrate genome annotation (Vega) database

The Vertebrate Genome Annotation (Vega) database (http://vega.sanger.ac.uk) was first made public in 2004 and has been designed to view manual annotation of human, mouse and zebrafish genomic sequences produced at the Wellcome Trust Sanger Institute. Since its initial release, the number of human annotated loci has more than doubled to close to 33 000 and now contains comprehensive annotation on 20 of the 24 human chromosomes, four whole mouse chromosomes and around 40% of the zebrafish Danio rerio genome. In addition, we offer manual annotation of a number of haplotype regions in mouse and human and regions of comparative interest in pig and dog that are unique to Vega.

The DNA sequence and biological annotation of human chromosome 1

The reference sequence for each human chromosome provides the framework for understanding genome function, variation and evolution. Here we report the finished sequence and biological annotation of human chromosome 1. Chromosome 1 is gene-dense, with 3,141 genes and 991 pseudogenes, and many coding sequences overlap. Rearrangements and mutations of chromosome 1 are prevalent in cancer and many other diseases. Patterns of sequence variation reveal signals of recent selection in specific genes that may contribute to human fitness, and also in regions where no function is evident. Fine-scale recombination occurs in hotspots of varying intensity along the sequence, and is enriched near genes. These and other studies of human biology and disease encoded within chromosome 1 are made possible with the highly accurate annotated sequence, as part of the completed set of chromosome sequences that comprise the reference human genome.

The Vertebrate Genome Annotation (Vega) database

The Vertebrate Genome Annotation (Vega) database (http://vega.sanger.ac.uk) has been designed to be a community resource for browsing manual annotation of finished sequences from a variety of vertebrate genomes. Its core database is based on an Ensembl-style schema, extended to incorporate curation-specific metadata. In collaboration with the genome sequencing centres, Vega attempts to present consistent high-quality annotation of the published human chromosome sequences. In addition, it is also possible to view various finished regions from other vertebrates, including mouse and zebrafish. Vega displays only manually annotated gene structures built using transcriptional evidence, which can be examined in the browser. Attempts have been made to standardize the annotation procedure across each vertebrate genome, which should aid comparative analysis of orthologues across the different finished regions.

Ensembl 2005

The Ensembl (http://www.ensembl.org/) project provides a comprehensive and integrated source of annotation of large genome sequences. Over the last year the number of genomes available from the Ensembl site has increased by 7 to 16, with the addition of the six vertebrate genomes of chimpanzee, dog, cow, chicken, tetraodon and frog and the insect genome of honeybee. The majority have been annotated automatically using the Ensembl gene build system, showing its flexibility to reliably annotate a wide variety of genomes. With the increased number of vertebrate genomes, the comparative analysis provided to users has been greatly improved, with new website interfaces allowing annotation of different genomes to be directly compared. The Ensembl software system is being increasingly widely reused in different projects showing the benefits of a completely open approach to software development and distribution.

DNA sequence and analysis of human chromosome 9

Chromosome 9 is highly structurally polymorphic. It contains the largest autosomal block of heterochromatin, which is heteromorphic in 6-8% of humans, whereas pericentric inversions occur in more than 1% of the population. The finished euchromatic sequence of chromosome 9 comprises 109,044,351 base pairs and represents >99.6% of the region. Analysis of the sequence reveals many intra- and interchromosomal duplications, including segmental duplications adjacent to both the centromere and the large heterochromatic block. We have annotated 1,149 genes, including genes implicated in male-to-female sex reversal, cancer and neurodegenerative disease, and 426 pseudogenes. The chromosome contains the largest interferon gene cluster in the human genome. There is also a region of exceptionally high gene and G + C content including genes paralogous to those in the major histocompatibility complex. We have also detected recently duplicated genes that exhibit different rates of sequence divergence, presumably reflecting natural selection.

The DNA sequence and comparative analysis of human chromosome 10

The finished sequence of human chromosome 10 comprises a total of 131,666,441 base pairs. It represents 99.4% of the euchromatic DNA and includes one megabase of heterochromatic sequence within the pericentromeric region of the short and long arm of the chromosome. Sequence annotation revealed 1,357 genes, of which 816 are protein coding, and 430 are pseudogenes. We observed widespread occurrence of overlapping coding genes (either strand) and identified 67 antisense transcripts. Our analysis suggests that both inter- and intrachromosomal segmental duplications have impacted on the gene count on chromosome 10. Multispecies comparative analysis indicated that we can readily annotate the protein-coding genes with current resources. We estimate that over 95% of all coding exons were identified in this study. Assessment of single base changes between the human chromosome 10 and chimpanzee sequence revealed nonsense mutations in only 21 coding genes with respect to the human sequence.

The DNA sequence and analysis of human chromosome 13

Chromosome 13 is the largest acrocentric human chromosome. It carries genes involved in cancer including the breast cancer type 2 (BRCA2) and retinoblastoma (RB1) genes, is frequently rearranged in B-cell chronic lymphocytic leukaemia, and contains the DAOA locus associated with bipolar disorder and schizophrenia. We describe completion and analysis of 95.5 megabases (Mb) of sequence from chromosome 13, which contains 633 genes and 296 pseudogenes. We estimate that more than 95.4% of the protein-coding genes of this chromosome have been identified, on the basis of comparison with other vertebrate genome sequences. Additionally, 105 putative non-coding RNA genes were found. Chromosome 13 has one of the lowest gene densities (6.5 genes per Mb) among human chromosomes, and contains a central region of 38 Mb where the gene density drops to only 3.1 genes per Mb.

Ensembl 2004

The Ensembl (http://www.ensembl.org/) database project provides a bioinformatics framework to organize biology around the sequences of large genomes. It is a comprehensive and integrated source of annotation of large genome sequences, available via interactive website, web services or flat files. As well as being one of the leading sources of genome annotation, Ensembl is an open source software engineering project to develop a portable system able to handle very large genomes and associated requirements. The facilities of the system range from sequence analysis to data storage and visualization and installations exist around the world both in companies and at academic sites. With a total of nine genome sequences available from Ensembl and more genomes to follow, recent developments have focused mainly on closer integration between genomes and external data.

The DNA sequence and analysis of human chromosome 6

Chromosome 6 is a metacentric chromosome that constitutes about 6% of the human genome. The finished sequence comprises 166,880,988 base pairs, representing the largest chromosome sequenced so far. The entire sequence has been subjected to high-quality manual annotation, resulting in the evidence-supported identification of 1,557 genes and 633 pseudogenes. Here we report that at least 96% of the protein-coding genes have been identified, as assessed by multi-species comparative sequence analysis, and provide evidence for the presence of further, otherwise unsupported exons/genes. Among these are genes directly implicated in cancer, schizophrenia, autoimmunity and many other diseases. Chromosome 6 harbours the largest transfer RNA gene cluster in the genome; we show that this cluster co-localizes with a region of high transcriptional activity. Within the essential immune loci of the major histocompatibility complex, we find HLA-B to be the most polymorphic gene on chromosome 6 and in the human genome.

Ensembl 2002: accommodating comparative genomics

The Ensembl (http://www.ensembl.org/) database project provides a bioinformatics framework to organise biology around the sequences of large genomes. It is a comprehensive source of stable automatic annotation of human, mouse and other genome sequences, available as either an interactive web site or as flat files. Ensembl also integrates manually annotated gene structures from external sources where available. As well as being one of the leading sources of genome annotation, Ensembl is an open source software engineering project to develop a portable system able to handle very large genomes and associated requirements. These range from sequence analysis to data storage and visualisation and installations exist around the world in both companies and at academic sites. With both human and mouse genome sequences available and more vertebrate sequences to follow, many of the recent developments in Ensembl have focusing on developing automatic comparative genome analysis and visualisation.

The Ensembl genome database project

The Ensembl (http://www.ensembl.org/) database project provides a bioinformatics framework to organise biology around the sequences of large genomes. It is a comprehensive source of stable automatic annotation of the human genome sequence, with confirmed gene predictions that have been integrated with external data sources, and is available as either an interactive web site or as flat files. It is also an open source software engineering project to develop a portable system able to handle very large genomes and associated requirements from sequence analysis to data storage and visualisation. The Ensembl site is one of the leading sources of human genome sequence annotation and provided much of the analysis for publication by the international human genome project of the draft genome. The Ensembl system is being installed around the world in both companies and academic sites on machines ranging from supercomputers to laptops.

The DNA sequence and comparative analysis of human chromosome 20

The finished sequence of human chromosome 20 comprises 59,187,298 base pairs (bp) and represents 99.4% of the euchromatic DNA. A single contig of 26 megabases (Mb) spans the entire short arm, and five contigs separated by gaps totalling 320 kb span the long arm of this metacentric chromosome. An additional 234,339 bp of sequence has been determined within the pericentromeric region of the long arm. We annotated 727 genes and 168 pseudogenes in the sequence. About 64% of these genes have a 5' and a 3' untranslated region and a complete open reading frame. Comparative analysis of the sequence of chromosome 20 to whole-genome shotgun-sequence data of two other vertebrates, the mouse Mus musculus and the puffer fish Tetraodon nigroviridis, provides an independent measure of the efficiency of gene annotation, and indicates that this analysis may account for more than 95% of all coding exons and almost all genes.

The Interdisciplinary Generalist Curriculum Project at Eastern Virginia Medical School

The proposed Interdisciplinary Generalist Curriculum (IGC) Project at Eastern Virginia Medical School (hereafter Eastern Virginia) intended to encourage students to select generalist disciplines by featuring generalist role models, focusing on patients' perspectives, teaching generalist skills early, providing care to indigent and other populations, and emphasizing students' personal and professional development. To do so, Eastern Virginia proposed that collaborative interdisciplinary groups of faculty plan and oversee the implementation of first- and second-year students' early clinical experiences in generalists' offices as integrated with new and revised first- and second-year courses, the coordination of generalist curricula longitudinally from year one through year four, and the provision of appropriate faculty development. With minor exceptions described, the project was implemented as proposed. The project did have desirable effects, both intended and unexpected. The curricular changes made in the project will remain.

Initial sequencing and analysis of the human genome

The human genome holds an extraordinary trove of information about human development, physiology, medicine and evolution. Here we report the results of an international collaboration to produce and make freely available a draft sequence of the human genome. We also present an initial analysis of the data, describing some of the insights that can be gleaned from the sequence.

Analysis and assessment of ab initio three-dimensional prediction, secondary structure, and contacts prediction

CASP3 saw a substantial increase in the volume of ab initio 3D prediction data, with 507 datasets for fifteen selected targets and sixty-one groups participating. As with CASP2, methods ranged from computationally intensive strategies that attempt to recreate the physical and chemical forces involved in protein folding to the more recent knowledge-based approaches. These exploit information from the structure databases, extracting potentially similar fragments and/or distance constraints derived from multiple sequence alignments. The knowledge-based approaches generally gave more consistently successful predictions across the range of targets, particularly that of the Baker group (Bystroff and Baker, J Mol Biol 1998;281:565-577; Simons et al. Proteins Suppl 1999;3:171-176), which used a fragment library. In the secondary structure prediction category, the most successful approaches built on the concepts used in PHD (Rost et al. Comput Appl Biosci 1994;10:53-60), an accepted standard in this field. Like PHD, they exploit neural networks but have different strategies for incorporating multiple sequence data or position-dependent weight matrices for training the networks. Analysis of the contact data, for which only six groups participated, suggested that as yet this data provides a rather weak signal. However, in combination with other types of prediction data it can sometimes be a useful constraint for identifying the correct structure.

SPEM: a parser for EMBL style flat file database entries

SUMMARY

We present a set of Perl modules for the flexible and robust parsing and editing of EMBL/SWISS-PROT databases.

AVAILABILITY

The Web page at http://www.sanger.ac. uk/Software/PerlModule/ provides information about downloading the SPEM and PrEMBL modules, and provides links to documentation and example code.

Sequence comparisons using multiple sequences detect three times as many remote homologues as pairwise methods

The sequences of related proteins can diverge beyond the point where their relationship can be recognised by pairwise sequence comparisons. In attempts to overcome this limitation, methods have been developed that use as a query, not a single sequence, but sets of related sequences or a representation of the characteristics shared by related sequences. Here we describe an assessment of three of these methods: the SAM-T98 implementation of a hidden Markov model procedure; PSI-BLAST; and the intermediate sequence search (ISS) procedure. We determined the extent to which these procedures can detect evolutionary relationships between the members of the sequence database PDBD40-J. This database, derived from the structural classification of proteins (SCOP), contains the sequences of proteins of known structure whose sequence identities with each other are 40% or less. The evolutionary relationships that exist between those that have low sequence identities were found by the examination of their structural details and, in many cases, their functional features. For nine false positive predictions out of a possible 432,680, i.e. at a false positive rate of about 1/50,000, SAM-T98 found 35% of the true homologous relationships in PDBD40-J, whilst PSI-BLAST found 30% and ISS found 25%. Overall, this is about twice the number of PDBD40-J relations that can be detected by the pairwise comparison procedures FASTA (17%) and GAP-BLAST (15%). For distantly related sequences in PDBD40-J, those pairs whose sequence identity is less than 30%, SAM-T98 and PSI-BLAST detect three times the number of relationships found by the pairwise methods.

Using neural networks for prediction of the subcellular location of proteins

Neural networks have been trained to predict the subcellular location of proteins in prokaryotic or eukaryotic cells from their amino acid composition. For three possible subcellular locations in prokaryotic organisms a prediction accuracy of 81% can be achieved. Assigning a reliability index, 33% of the predictions can be made with an accuracy of 91%. For eukaryotic proteins (excluding plant sequences) an overall prediction accuracy of 66% for four locations was achieved, with 33% of the sequences being predicted with an accuracy of 82% or better. With the subcellular location restricting a protein's possible function, this method should be a useful tool for the systematic analysis of genome data and is available via a server on the world wide web.

GLASS: a tool to visualize protein structure prediction data in three dimensions and evaluate their consistency

When a protein sequence does not share any significant sequence similarity with a protein of known structure, homology modeling cannot be applied. However, many novel and interesting methods, such as secondary structure prediction, fold recognition, and prediction of long-range interactions, are being developed and have been shown to be reasonably successful in predicting protein structures from sequence data and evolutionary information. The a priori evaluation of the correctness of a prediction obtained by one of these methods is however often problematic. Consequently, it is important to use all available information provided by as many different methods as possible and all the available experimental data about the protein of interest, since the consistency of the results is indicative of the reliability of the prediction. Hence the need has arisen for suitable tools able to compare results provided by different methods and evaluate their consistency. We have therefore constructed GLASS, a general platform to read, visualize, compare, and evaluate prediction results from many different sources and to project these prediction results into three dimensions. In addition, GLASS allows the comparison of selected parameters calculated for a model with the distribution observed in real protein structures, thus providing an easy way to test new methods for evaluating the likelihood of different structural models. GLASS can be considered as a "workbench" for structural predictions useful to both experimentalists and theoreticians.

Intermediate sequences increase the detection of homology between sequences

Two homologous sequences, which have diverged beyond the point where their homology can be recognised by a simple direct comparison, can be related through a third sequence that is suitably intermediate between the two. High scores, for a sequence match between the first and third sequences and between the second and the third sequences, imply that the first and second sequences are related even though their own match score is low. We have tested the usefulness of this idea using a database that contains the sequences of 971 protein domains whose structures are known and whose residue identities with each other are some 40% or less (PDB40D). On the basis of sequence and structural information, 2143 pairs of these sequences are known to have an evolutionary relationship. FASTA, in an all-against-all comparison of the sequences in the database, detected 320 (15%) of these relationships as well as three false positive (i.e. 1% error rate). Using intermediate sequences found by FASTA matches of PDB40D sequences to those in the large non-redundant OWL database we could detect 550 evolutionary relationships with an error rate of 1%. This means the intermediate sequence procedure increases the ability to recognise the evolutionary relationships amongst the PDB40D sequences by 70%.

Protein folds in the all-beta and all-alpha classes

Analysis of the structures in the Protein Databank, released in June 1996, shows that the number of different protein folds, i.e. the number of different arrangements of major secondary structures and/or chain topologies, is 327. Of these folds, approximately 25% belong to the all-alpha class, 20% belong to the all-beta class, 30% belong to the alpha/beta class, and 25% belong to the alpha + beta class. We describe the types of folds now known for the all-beta and all-alpha classes, emphasizing those that have been discovered recently. Detailed theories for the physical determinants of the structures of most of these folds now exist, and these are reviewed.

Update on protein structure prediction: results of the 1995 IRBM workshop

Computational tools for protein structure prediction are of great interest to molecular, structural and theoretical biologists due to a rapidly increasing number of protein sequences with no known structure. In October 1995, a workshop was held at IRBM to predict as much as possible about a number of proteins of biological interest using ab initio prediction of fold recognition methods. 112 protein sequences were collected via an open invitation for target submissions. 17 were selected for prediction during the workshop and for 11 of these a prediction of some reliability could be made. We believe that this was a worthwhile experiment showing that the use of a range of independent prediction methods and thorough use of existing databases can lead to credible and useful ab initio structure predictions.

Discovery of an orally bioavailable NK1 receptor antagonist, (2S,3S)-(2-methoxy-5-tetrazol-1-ylbenzyl)(2-phenylpiperidin-3-yl)amine (GR203040), with potent antiemetic activity

The antiemetic, pharmacokinetic, and metabolic profile of CP-99,994, a potent NK1 receptor antagonist, has been carefully evaluated. As a result we began a medicinal chemistry program which initially identified a 3-furanyl analogue (6) with improved antiemetic potency and a methyl sulfone (5) with enhanced metabolic stability and oral bioavailability. The improved pharmacokinetic profile of methyl sulfone (5) was associated with its low lipophilicity, and a therefore a number of heterocyclic analogues with reduced log D were synthesized. Out of this program emerged 19 (GR203040), a tetrazolyl-substituted analogue. Tetrazole 19 inhibits radiation-induced emesis in the ferret with high potency when administered both subcutaneously and orally, has a long duration of action, and has high oral bioavailability in the dog. Tetrazole 19 is currently undergoing evaluation as a novel approach for the control of emesis associated with, for example, cancer chemotherapy.

SCOP: a structural classification of proteins database for the investigation of sequences and structures

To facilitate understanding of, and access to, the information available for protein structures, we have constructed the Structural Classification of Proteins (scop) database. This database provides a detailed and comprehensive description of the structural and evolutionary relationships of the proteins of known structure. It also provides for each entry links to co-ordinates, images of the structure, interactive viewers, sequence data and literature references. Two search facilities are available. The homology search permits users to enter a sequence and obtain a list of any structures to which it has significant levels of sequence similarity. The key word search finds, for a word entered by the user, matches from both the text of the scop database and the headers of Brookhaven Protein Databank structure files. The database is freely accessible on World Wide Web (WWW) with an entry point to URL http: parallel scop.mrc-lmb.cam.ac.uk magnitude of scop.

Modified crush-avulsion anastomosis model on the rat femoral vein

A crush-avulsion anastomosis model was established on the rat femoral vein. Saline or heparin was used as a luminal topical agent and was allowed to contact the damaged endothelium for 10 min. Patency and coagulation parameters were investigated for 1 week. The heparin treated group had a patency rate of 93% at 1 hr vs. 13% for the saline treated group (P < 0.001). At 7 days, the heparin treated group had an 87% patency vs. 7% for the saline-treated group (P < 0.001). Scanning electron micrography (SEM) provided evidence of the deposition of the components of early thrombosis in the crushed venous wall. In contrast, the SEM of the heparin treated group shows a paucity of any evidence of thrombus. These results indicate that the rat vein crush-avulsion model is a reliable and reproducible thrombosis model with low patency. The methods used with the topical agent may improve the patency rate in crush avulsion injuries.

Complex cranial base trauma resulting from recreational fireworks injury: case reports and review of the literature

Two patients who sustained complex skull base trauma secondary to recreational fireworks injuries are reported. Initial assessment and management included axial and coronal computerized tomography, control of hemorrhage, debridement of wound and brain, isolation of brain from external environment, and reconstruction of the cranial base floor. Secondary orbital and facial reconstruction used available bone fragments and iliac bone graft in one patient and vascularized free tissue transfer in the other. In both patients, reconstruction of both the intracranial and extracranial compartments was successful with acceptable cosmetic result. Modification of multiple conventional approaches, along with a multispecialty surgical team, was used to deal effectively with these unique cases.

Use of recombinant hirudin to increase the patency rate of microanastomosis in a rabbit model

Microvascular surgical techniques now play an extremely important role in reconstructive surgery. Failures still persist and are most commonly due to thrombotic occlusion of the microvascular anastomosis. Five to 20% of all free tissue transfers and replants are affected. A logical step in the effort to resolve this problem would be the infusion of the newly available recombinant hirudin (r-hirudin). The value of r-hirudin in microvascular surgery has not been investigated. An animal model mimicking the thrombotic mechanisms similar to those in the clinical situation has been described in which a length of rabbit femoral artery is excised, turned inside out and reinserted into the vessel with two standard microvascular anastomoses. The major factors in clinical anastomosis failure are represented in this animal model by medial disruption, imperfect vessel wall alignment, needle holes, trauma, foreign suture material and changes in flow. The type III adventitial collagen turned inward reproduces the clinical problem of adventitial strands becoming caught in the anastomosis. A predictable rate of thrombosis results depending on inversion graft length: a 2 mm segment has an occlusion rate at one week of about 30%. Using this model heparin infusions significantly decrease thrombosis rates compared to saline. Using an almost identical model we have begun a study the effect of r-hirudin on microvascular patency. New Zealand White male rabbits were anaesthetized with ketamine and xylazine. Through a groin incision the femoral vein was catheterized and tubing connected to an infusion pump. The femoral artery was exposed, and the adventitia aggressively removed since it is difficult to minimally debride adventitia in a reproducible fashion.(ABSTRACT TRUNCATED AT 250 WORDS)

Protein engineering and design

Rapid advances in site-directed mutagenesis and total gene synthesis combined with new expression systems in prokaryotic and eukaryotic cells have provided the molecular biologist with tools for modification of existing proteins to improve catalytic activity, stability and selectivity, for construction of chimeric molecules and for synthesis of completely novel molecules that may be endowed with some useful activity. Such protein engineering can be seen as a cycle in which the structures of engineered molecules are studied by X-ray analysis and two-dimensional nuclear magnetic resonance. The results are used in the improvement of the design by using knowledge-based procedures that exploit facts, rules and observations about proteins of known three-dimensional structure.

18th Sir Hans Krebs lecture. Knowledge-based protein modelling and design

A systematic technique for protein modelling that is applicable to the design of drugs, peptide vaccines and novel proteins is described. Our approach is knowledge-based, depending on the structures of homologous or analogous proteins and more generally on a relational data base of protein three-dimensional structures. The procedure simultaneously aligns the known tertiary structures, selects fragments from the structurally conserved regions on the basis of sequence homology, aligns these with the 'average structure' or 'framework', builds on the loops selected from homologous proteins or a wider database, substitutes sidechains and energy minimises the resultant model. Applications to modelling an homologous structure, tissue plasminogen activator on the basis of another serine proteinase, and to modelling an analogous protein, HIV viral proteinase on the basis of aspartic proteinases, are described. The converse problem of ab initio design is also addressed: this involves the selection of an amino acid sequence to give a particular tertiary structure, in this case a symmetrical domain of two Greek-key motifs.