Classification and regression trees are machine-learning methods for constructing prediction models from data. The models are obtained by recursively partitioning the data space and fitting a simple prediction model within each partition. As a result, the partitioning can be represented graphically as a decision tree. Classification trees are designed for dependent variables that take a finite number of unordered values, with prediction error measured in terms of misclassification cost. Regression trees are for dependent variables that take continuous or ordered discrete values, with prediction error typically measured by the squared difference between the observed and predicted values. This article gives an introduction to the subject by reviewing some widely available algorithms and comparing their capabilities, strengths, and weakness in two examples. © 2011 John Wiley & Sons, Inc. WIREs Data Mining Knowl Discov 2011 1 14-23 DOI: 10.1002/widm.8

This article is categorized under: 

Technologies > Classification
Technologies > Machine Learning
Technologies > Prediction
Technologies > Statistical Fundamentals

/pdf/classification-and-regression-trees-1td8f8b4ru.pdf

Classification and regression trees

RNA interference (RNAi) is the process of sequence-specific, post-transcriptional gene silencing in animals and plants, initiated by double-stranded RNA (dsRNA) that is homologous in sequence to the silenced gene. The mediators of sequence-specific messenger RNA degradation are 21- and 22-nucleotide small interfering RNAs (siRNAs) generated by ribonuclease III cleavage from longer dsRNAs. Here we show that 21-nucleotide siRNA duplexes specifically suppress expression of endogenous and heterologous genes in different mammalian cell lines, including human embryonic kidney (293) and HeLa cells. Therefore, 21-nucleotide siRNA duplexes provide a new tool for studying gene function in mammalian cells and may eventually be used as gene-specific therapeutics.

Duplexes of 21-nucleotide RNAs mediate RNA interference in cultured mammalian cells

One of the mechanisms by which the innate immune system senses the invasion of pathogenic microorganisms is through the Toll-like receptors (TLRs), which recognize specific molecular patterns that are present in microbial components. Stimulation of different TLRs induces distinct patterns of gene expression, which not only leads to the activation of innate immunity but also instructs the development of antigen-specific acquired immunity. Here, we review the rapid progress that has recently improved our understanding of the molecular mechanisms that mediate TLR signalling.

Toll-like receptor signalling

The innate immune system in drosophila and mammals senses the invasion of microorganisms using the family of Toll receptors, stimulation of which initiates a range of host defense mechanisms. In drosophila antimicrobial responses rely on two signaling pathways: the Toll pathway and the IMD pathway. In mammals there are at least 10 members of the Toll-like receptor (TLR) family that recognize specific components conserved among microorganisms. Activation of the TLRs leads not only to the induction of inflammatory responses but also to the development of antigen-specific adaptive immunity. The TLR-induced inflammatory response is dependent on a common signaling pathway that is mediated by the adaptor molecule MyD88. However, there is evidence for additional pathways that mediate TLR ligand-specific biological responses.

Toll-like receptors.

Since the first application of the surface plasmon resonance (SPR) phenomenon for sensing almost two decades ago, this method has made great strides both in terms of instrumentation development and applications. SPR sensor technology has been commercialized and SPR biosensors have become a central tool for characterizing and quantifying biomolecular interactions. This paper attempts to review the major developments in SPR technology. Main application areas are outlined and examples of applications of SPR sensor technology are presented. Future prospects of SPR sensor technology are discussed.

Surface plasmon resonance sensors: review

The central event in the cellular immune response to invading microorganisms is the specific recognition of foreign peptides bound to major histocompatibility complex (MHC) molecules by the αβ T cell receptor (TCR). The x-ray structure of the complete extracellular fragment of a glycosylated αβ TCR was determined at 2.5 angstroms, and its orientation bound to a class I MHC-peptide (pMHC) complex was elucidated from crystals of the TCR-pMHC complex. The TCR resembles an antibody in the variable Vα and Vβ domains but deviates in the constant Cα domain and in the interdomain pairing of Cα with Cβ. Four of seven possible asparagine-linked glycosylation sites have ordered carbohydrate moieties, one of which lies in the Cα-Cβ interface. The TCR combining site is relatively flat except for a deep hydrophobic cavity between the hypervariable CDR3s (complementarity-determining regions) of the α and β chains. The 2C TCR covers the class I MHC H-2K b binding groove so that the Vα CDRs 1 and 2 are positioned over the amino-terminal region of the bound dEV8 peptide, the Vβ chain CDRs 1 and 2 are over the carboxyl-terminal region of the peptide, and the Vα and Vβ CDR3s straddle the peptide between the helices around the central position of the peptide.

https://science.sciencemag.org/content/sci/274/5285/209.full.pdf

An αβ T cell receptor structure at 2.5 Å and its orientation in the TCR-MHC complex

Several families of transmembrane endoplasmic reticulum (ER) proteins contain retention motifs in their cytoplasmically exposed tails. Mutational analyses demonstrated that two lysines positioned three and four or five residues from the C-terminus represent the retention motif. The introduction of a lysine preceding the lysine that occurs three residues from the terminus of Lyt2 renders this cell surface protein a resident of the ER. Likewise, the appropriate positioning of two lysine residues in a poly-serine sequence confines marker proteins to the ER. Arginines or histidines cannot replace lysines, suggesting that simple charge interactions are not sufficient to explain the retention. The identified consensus motif may serve as a retrieval signal that brings proteins back from a sorting compartment adjacent to the ER.

Identification of a consensus motif for retention of transmembrane proteins in the endoplasmic reticulum.

The x-ray structures of a murine MHC class I molecule (H-2Kb) were determined in complex with two different viral peptides, derived from the vesicular stomatitis virus nucleoprotein (52-59), VSV-8, and the Sendai virus nucleoprotein (324-332), SEV-9. The H-2Kb complexes were refined at 2.3 A for VSV-8 and 2.5 A for SEV-9. The structure of H-2Kb exhibits a high degree of similarity with human HLA class I, although the individual domains can have slightly altered dispositions. Both peptides bind in extended conformations with most of their surfaces buried in the H-2Kb binding groove. The nonamer peptide maintains the same amino- and carboxyl-terminal interactions as the octamer primarily by the insertion of a bulge in the center of an otherwise beta conformation. Most of the specific interactions are between side-chain atoms of H-2Kb and main-chain atoms of peptide. This binding scheme accounts in large part for the enormous diversity of peptide sequences that bind with high affinity to class I molecules. Small but significant conformational changes in H-2Kb are associated with peptide binding, and these synergistic movements may be an integral part of the T cell receptor recognition process.

https://science.sciencemag.org/content/sci/257/5072/919.full.pdf

Crystal structures of two viral peptides in complex with murine MHC class I H-2Kb.

β-Amyloid1–42 Binds to α7 Nicotinic Acetylcholine Receptor with High Affinity IMPLICATIONS FOR ALZHEIMER'S DISEASE PATHOLOGY

Class I major histocompatibility complex (MHC) molecules interact with self and foreign peptides of diverse amino acid sequences yet exhibit distinct allele-specific selectivity for peptide binding. The structures of the peptide-binding specificity pockets (subsites) in the groove of murine H-2Kb as well as human histocompatibility antigen class I molecules have been analyzed. Deep but highly conserved pockets at each end of the groove bind the amino and carboxyl termini of peptide through extensive hydrogen bonding and, hence, dictate the orientation of peptide binding. A deep polymorphic pocket in the middle of the groove provides the chemical and structural complementarity for one of the peptide's anchor residues, thereby playing a major role in allele-specific peptide binding. Although one or two shallow pockets in the groove may also interact with specific peptide side chains, their role in the selection of peptide is minor. Thus, usage of a limited number of both deep and shallow pockets in multiple combinations appears to allow the binding of a broad range of peptides. This binding occurs with high affinity, primarily because of extensive interactions with the peptide backbone and the conserved hydrogen bonding network at both termini of the peptide. Interactions between the anchor residue (or residues) and the corresponding allele-specific pocket provide sufficient extra binding affinity not only to enhance specificity but also to endure the presentation of the peptide at the cell surface for recognition by T cells.

http://science.sciencemag.org/content/sci/257/5072/927.full.pdf

Per A. Peterson

Papers

An αβ T cell receptor structure at 2.5 Å and its orientation in the TCR-MHC complex

Identification of a consensus motif for retention of transmembrane proteins in the endoplasmic reticulum.

Crystal structures of two viral peptides in complex with murine MHC class I H-2Kb.

β-Amyloid1–42 Binds to α7 Nicotinic Acetylcholine Receptor with High Affinity IMPLICATIONS FOR ALZHEIMER'S DISEASE PATHOLOGY

Emerging principles for the recognition of peptide antigens by MHC class I molecules