Evaluating Protein Motif Significance Measures: A Case Study on Prosite Patterns

In biology the notion of motif plays a central role for describing various phenomena. For example, protein functional motifs, like the ones contained in the PROSITE database [20], e.g. [FY ]DPC[LIM ][ASG]C[ASG] , are in general represented as motifs with character classes. These motifs are collected using semi-automatic procedures, nevertheless they are still manually verified. The discovery of sequence motifs in proteins and genes is becoming increasingly important [2, 29]. Such motifs usually correspond to residues conserved during evolution due to some significant structural or functional role. Moreover the increasing availability of biological sequences, such as whole genomes, from next-generation sequencing technologies to new protein discoveries, has increased the need for automatic methods for their analysis and comparison. In order to fill this gap, researchers have developed several approaches over the years. Typically these approaches follow in popular frameworks of motif discovery

VISA (VIsual Sequence Analysis) is a software package that displays global similarities within a set of related protein sequences. The program identifies amino acid patterns that are common to many members of the set of sequences and displays them as a series of histograms. Individual peaks on the display can be assigned a color and analogous peaks in the other sequences are then automatically marked in the same color. This can be repeated for each significant peak and leads to a display in which major matching segments of multiple amino acid sequences appear as dominant peaks of the histograms with matching colors. These peaks usually correspond to the conserved sequence motifs that are characteristic of particular proteins. An extensive set of software tools is included to help the localization, visualization and analysis of the global similarities displayed. VISA provides a graphic overview of the sequence similarity that can help to understand the architecture of the protein family and can be helpful while designing experiments to probe function.

This paper presents an essentially new method used to construct phylogenetic trees from related amino acid sequences. The method is based on a new distance measure which describes sequence relationships by means of typical steric and physicochemical properties of the amino acids and is advantageous in some essential points. The method was applied to different sets of protein sequences and the results were compared with other well-established methods.

We have developed a computer algorithm that can extract the pattern of conserved primary sequence elements common to all members of a homologous protein family. The method involves clustering the pairwise similarity scores among a set of related sequences to generate a binary dendrogram (tree). The tree is then reduced in a stepwise manner by progressively replacing the node connecting the two most similar termini by one common pattern until only a single common "root" pattern remains. A pattern is generated at a node by (i) performing a local optimal alignment on the sequence/pattern pair connected by the node with the use of an extended dynamic programming algorithm and then (ii) constructing a single common pattern from this alignment with a nested hierarchy of amino acid classes to identify the minimal inclusive amino acid class covering each paired set of elements in the alignment. Gaps within an alignment are created and/or extended using a "pay once" gap penalty rule, and gapped positions are converted into gap characters that function as 0 or 1 amino acid of any type during subsequent alignment. This method has been used to generate a library of covering patterns for homologous families in the National Biomedical Research Foundation/Protein Identification Resource protein sequence data base. We show that a covering pattern can be more diagnostic for sequence family membership than any of the individual sequences used to construct the pattern.

Biological pattern discovery problems are computationally expensive. A possible technique for reducing the time to perform pattern discovery is parallelization. Since each task in a biological pattern discovery application is usually time-consuming by itself, we might be able to use networks of workstations (NOWs) that communicate infrequently. Persistent Linda (PLinda) is a distributed parallel computing system that runs on NOWs and it automatically utilizes idle workstations (Anderson and Shasha, 1992; Jeong, 1996). This means that labs can do parallel pattern discovery without buying new hardware. We propose an acyclic directed graph structure, exploration dag (E-dag for short), to characterize computational models of biological pattern discovery applications. An E-dag can first be constructively formed from specifications of a pattern discovery problem; then an E-dag traversal is performed on the fly to solve the problem. When done in parallel, the process of E-dag construction and traversal efficiently solves pattern discovery problems. Parallel E-dag construction and traversal can be easily programmed in PLinda. Finding active motifs in sets of protein sequences and in multiple RNA secondary structures are two examples of biological pattern discovery. Before discussing the framework, we introduce these two applications and briefly describe their computational models. Consider a database of imaginary protein sequences D = {FFRR, MRRM, MTRM, DPKY, AVLG} and the query “Find the patterns P of the form *X* where P occurs in at least two sequences in D and the size of P |P| ≥ 2.” (X can be a segment of a sequence of any length, and * represents a variable length don’t care [VLDC].) The good patterns are *RR* (which occurs in FFRR and MRRM) and *RM* (which occurs in MRRM and MTRM). Pattern discovery in sets of sequences concerns finding commonly occurring subsequences (sometimes called motifs). The structures of the motifs we wish to discover are regular expressions of the form *S1 * S2 * ... where S1,S2,… are segments of a sequence, that is, subsequences made up of consecutive letters, and * represents a VLDC. In matching the expression *S1 * S2 * … with a sequence S, the VLDCs may substitute for zero or more letters in S.

The objective of the present study was to design a model for integrating process mining into the auditing of financial statements. The research employed an exploratory mixed-method approach (qualitative–quantitative). Participants in the qualitative phase included experts and specialists in financial statement auditing with experience in management and policy-making. Using purposive sampling and based on the principle of theoretical saturation, a sample size of 10 individuals was determined. Participants in the quantitative phase consisted of certified public accountants responsible for financial statement auditing in reporting companies in Tehran Province. Using stratified random sampling and Cochran’s formula, a sample size of 303 individuals was established. The data collection tools included semi-structured interviews for the qualitative phase and a researcher-made questionnaire for the quantitative phase. The qualitative data were analyzed using thematic analysis, and the quantitative data were analyzed through structural equation modeling with the use of SPSS and LISREL software. The findings indicated that 82% of the model design for integrating process mining into financial statement auditing depended on dimensions comprising four main themes: (1) individual factors, (2) environmental factors, (3) institutional factors, and (4) organizational factors; and four sub-themes: (1) adoption, (2) development of process mining in auditing, (3) implementation of process mining in auditing, and (4) performance evaluation of process mining in auditing. The results also demonstrated that the implementation of process mining enhances the reliability of audit results and, by replacing traditional manual auditing methods, strengthens the evidentiary power of audits. As a novel data mining technique, process mining equips auditors with tools to keep pace with technological advancements and challenges.

We present the Proteome Quality Index (PQI; http://pqi-list.org), a much-needed resource for users of bacterial and eukaryotic proteomes. Completely sequenced genomes for which there is an available set of protein sequences (the proteome) are given a one- to five-star rating supported by 11 different metrics of quality. The database indexes over 3000 proteomes at the time of writing and is provided via a website for browsing, filtering and downloading. Previous to this work, there was no systematic way to account for the large variability in quality of the thousands of proteomes, and this is likely to have profoundly influenced the outcome of many published studies, in particular large-scale comparative analyses. The lack of a measure of proteome quality is likely due to the difficulty in producing one, a problem that we have approached by integrating multiple metrics. The continued development and improvement of the index will require the contribution of additional metrics by us and by others; the PQI provides a useful point of reference for the scientific community, but it is only the first step towards a 'standard' for the field.

We propose an automated protocol for designing the energy landscape suitable for the description of a given set of protein sequences with known structures, by optimizing the parameters of the energy function. The parameters are optimized so that not only the global minimum-energy conformation becomes native-like, but also the conformations distinct from the native structure have higher energies than those close to the native one, for each protein sequence in the set. In order to achieve this goal, one has to sample protein conformations that are local minima of the energy function for given parameters. Then the parameters are optimized using linear approximation, and then local minimum conformations are searched with the new energy parameters. We develop an algorithm that repeats this process of parameter optimization based on linear approximation, and conformational optimization for the current parameters, ultimately leading to the optimization of the energy parameters. We test the feasibility of this algorithm by optimizing a coarse grained energy function, called the UNRES energy function, for a set of ten proteins.KeywordsEnergy FunctionStructural DatabaseProtein Structure PredictionConformational SearchAutomate ProtocolThese keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.

Peptide compositions constructed out of whole sets of protein sequences can be used as species signatures for phylogenetic analysis. To account for point mutations, an amino acid substitution model is integrated into the complete composition vectors through a novel peptide clustering algorithm. Such a refined signature is expected to highlight deeper evolutionary relationships among the species and employed into the whole genome phylogenetic analysis to define a new evolutionary distance measure. Computational experiments have been set up to validate the effectiveness of this new measure and a vertebrate evolutionary tree using a dataset of 832 proteins for 64 vertebrates is reported.

Fitness landscape analysis (FLA) is a technique to determine the characteristics of a problem or its structural features based on which the most appropriate algorithm is possible to recommend for solving the problem. In this chapter, we determine structural features of the protein structure prediction problem by analyzing the landscape structure. A landscape of the protein instances is generated by using the quasi-random sampling technique and city block distance. Structural features of the PSP Landscape are determined by applying various landscape measures. Numerical results indicate that the complexity of the PSP problem increases with protein sequence length. Six well-known real-coded optimization algorithms are evaluated over the same set of protein sequences and the performances are subsequently analyzed based on the structural features. Finally, we suggest the most appropriate algorithm(s) for solving the PSP problem.

Using Teiresias, a pattern discovery method that identifies all motifs present in any given set of protein sequences without requiring alignment or explicit enumeration of the solution space, we have explored the GenPept sequence database and built a dictionary of all sequence patterns with two or more instances. The entries of this dictionary, henceforth named seqlets, cover 98.12% of all amino acid positions in the input database and in essence provide a comprehensive finite set of descriptors for protein sequence space. As such, seqlets can be effectively used to describe almost every naturally occurring protein. In fact, seqlets can be thought of as building blocks of protein molecules that are a necessary (but not sufficient) condition for function or family equivalence memberships. Thus, seqlets can either define conserved family signatures or cut across molecular families and previously undetected sequence signals deriving from functional convergence. Moreover, we show that seqlets also can capture structurally conserved motifs. The availability of a dictionary of seqlets that has been derived in such an unsupervised, hierarchical manner is generating new opportunities for addressing problems that range from reliable classification and the correlation of sequence fragments with functional categories to faster and sensitive engines for homology searches, evolutionary studies, and protein structure prediction.

In this article, we propose an effective data mining technique for multi-class protein sequence classification. The technique, which can discover discriminative motif-sets for classification, performs its tasks in two phases. In Phase 1, it makes use of a popular motif discovery algorithm called MEME (Multiple Expectation Maximization for Motif Elicitation) to discover a set of highly conserved motifs in each protein family of training sequences. The highly conserved motif-sets discovered in each family may overlap with each other and may therefore not be unique enough to allow them to be used for classification. Phase 2, therefore, makes use of a pattern discovery approach to discover the interesting motif-sets in each protein family that are useful for classification with a single classifier. Based on these motif-sets, the functional family of each independent testing sequence can then be determined. For experimentation, the proposed technique has been tested with different sets of protein sequences. Experimental results show that it outperforms other existing protein sequence classifiers and can effectively classify proteins into their corresponding functional families. In addition, the motif-sets discovered during the training process have been found to be biologically meaningful.

The recent increase in the number of complete genetic sequences freely available through specialized Internet databases presents big challenges for the research community. One such challenge is the efficient and effective search of sequence patterns, also known as motifs, among a set of related genetic sequences. Such patterns describe regions that may provide important insights about the structural and functional role of DNA and proteins. Two main classes can be considered: probabilistic patterns represent a model that simulates the sequences or part of the sequences under consideration and deterministic patterns that either match or not the input sequences. In this chapter a general overview of deterministic sequence mining over sets of genetic sequences is proposed. The authors formulate an architecture that divides the mining process workflow into a set of blocks. Each of these blocks is discussed individually.

G protein-coupled receptors (GPCRs) contain a number of conserved proline residues in their transmembrane helices, and it is generally assumed these play important functional and/or structural roles. Here we use unnatural amino acid mutagenesis, employing α-hydroxy acids and proline analogues, to examine the functional roles of five proline residues in the transmembrane helices of the D2 dopamine receptor. The well-known tendency of proline to disrupt helical structure is important at all sites, while we find no evidence for a functional role for backbone amide cis-trans isomerization, another feature associated with proline. At most proline sites, the loss of the backbone NH is sufficient to explain the role of the proline. However, at one site, P210(5.50), a substituent on the backbone N appears to be essential for proper function. Interestingly, the pattern in functional consequences that we see is mirrored in the pattern of structural distortions seen in recent GPCR crystal structures.

In extravasation cancer cells and vascular niche are involved in a tight cross-talk which has been defined as the rate-regulating event for metastases establishment [1]. Recent, animal studies support the hypothesis that metastatic deficiency lies in focal adhesion complex alterations, however, it still needs to be elucidated which are the specific regulators of each event composing extravasation. Focal adhesion proteins Talin-1 (TLN-1) and Focal Adhesion Kinase (FAK) are up-regulated in breast cancer. Both targets due to their structural and functional role, dramatically influence cancer mechanotransduction leading to endothelial junction disruption, critical in extravasation process [2]. Here, we generated ad hoc engineered models for each extravasation step allowing single cell behavior analyses through high resolution real time imaging in a reliable and quantitative way in a physiological environment. Through this novel approach we analyzed the effect of TLN-1 and FAK and their genetic and chemical inhibition in breast cancer extravasation. The 3D-microfluidic vasculature displayed maturation markers and physiological permeability (1.5±0.76*10-6cm/s) and allowed cancer cell injection through the hollow vessels. Western blot confirmed TLN-1 and FAK knock down (KD) in MDA-MB231. Both targets significantly affected morphology and proliferation (p<0.001). We demonstrated the involvement of both targets in adhesion to the endothelium (p<0.001). Trans endothelial migration (TEM) was decreased in TLN-1 KD (p<0.05) and in FAK KD (P<0.01). Both TLN-1 and FAK KD cells were trapped into the vessels and were not able to extend pseudopodia. Cancer migration tracking in 3D matrix was statistically lower in both KD (p<0.001). The calculated ratios of adhesion/TEM and invasion/TEM were lower in TLN-1 compared to FAK KD, demonstrating that TLN-1 plays a major role in adhesion to endothelium and early invasion while FAK was identified as the main driver in TEM. Additive (TLN-1 and FAK) KD did not show significant difference respect to the target that mostly affected adhesion and invasion demonstrating TLN-1 role in these events. Additive KD was statistically lower compared to TLN-1 KD (P<0.05), supporting FAK crucial role in TEM. Inhibition of TLN-1 and FAK phosphorylation revealed that actin polymerization and pseudopodia formation (required for extravasation) were dependent by their structural role and were independent by their phosphorylation. However, the inhibition of FAK activity showed FAK phosphorylation as the key driver of TEM mechanism (p<0.01). Concluding, our results supported by in vivo data (p<0.05) showed that TLN-1 and FAK inhibition may represent novel strategies challenging mechanisms leading to metastatic establishment. 1. Cell,2011.147(2):p.275-92; 2. Proc Natl Acad Sci USA,2015.112(1):p.214-9 Citation Format: Mara Gilardi, Simone Bersini, Rosa Maria Moresco, Marco Vanoni, Roger D. Kamm, Matteo Moretti. Engineered microfluidic 3D human microvasculature identifies Talin-1-dependent adhesion and FAK activation as the key promoter of cancer cell trans-endothelial migration [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2017; 2017 Apr 1-5; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2017;77(13 Suppl):Abstract nr 5814. doi:10.1158/1538-7445.AM2017-5814