Search tips
Search criteria

Results 1-4 (4)

Clipboard (0)

Select a Filter Below

Year of Publication
Document Types
1.  Gauging NOTCH1 Activation in Cancer Using Immunohistochemistry 
PLoS ONE  2013;8(6):e67306.
Fixed, paraffin-embedded (FPE) tissues are a potentially rich resource for studying the role of NOTCH1 in cancer and other pathologies, but tests that reliably detect activated NOTCH1 (NICD1) in FPE samples have been lacking. Here, we bridge this gap by developing an immunohistochemical (IHC) stain that detects a neoepitope created by the proteolytic cleavage event that activates NOTCH1. Following validation using xenografted cancers and normal tissues with known patterns of NOTCH1 activation, we applied this test to tumors linked to dysregulated Notch signaling by mutational studies. As expected, frequent NICD1 staining was observed in T lymphoblastic leukemia/lymphoma, a tumor in which activating NOTCH1 mutations are common. However, when IHC was used to gauge NOTCH1 activation in other human cancers, several unexpected findings emerged. Among B cell tumors, NICD1 staining was much more frequent in chronic lymphocytic leukemia than would be predicted based on the frequency of NOTCH1 mutations, while mantle cell lymphoma and diffuse large B cell lymphoma showed no evidence of NOTCH1 activation. NICD1 was also detected in 38% of peripheral T cell lymphomas. Of interest, NICD1 staining in chronic lymphocytic leukemia cells and in angioimmunoblastic lymphoma was consistently more pronounced in lymph nodes than in surrounding soft tissues, implicating factors in the nodal microenvironment in NOTCH1 activation in these diseases. Among carcinomas, diffuse strong NICD1 staining was observed in 3.8% of cases of triple negative breast cancer (3 of 78 tumors), but was absent from 151 non-small cell lung carcinomas and 147 ovarian carcinomas. Frequent staining of normal endothelium was also observed; in line with this observation, strong NICD1 staining was also seen in 77% of angiosarcomas. These findings complement insights from genomic sequencing studies and suggest that IHC staining is a valuable experimental tool that may be useful in selection of patients for clinical trials.
PMCID: PMC3688991  PMID: 23825651
2.  Applying Small-Scale DNA Signatures as an Aid in Assembling Soybean Chromosome Sequences 
Advances in Bioinformatics  2010;2010:976792.
Previous work has established a genomic signature based on relative counts of the 16 possible dinucleotides. Until now, it has been generally accepted that the dinucleotide signature is characteristic of a genome and is relatively homogeneous across a genome. However, we found some local regions of the soybean genome with a signature differing widely from that of the rest of the genome. Those regions were mostly centromeric and pericentromeric, and enriched for repetitive sequences. We found that DNA binding energy also presented large-scale patterns across soybean chromosomes. These two patterns were helpful during assembly and quality control of soybean whole genome shotgun scaffold sequences into chromosome pseudomolecules.
PMCID: PMC2933861  PMID: 20827309
3.  Use of machine learning algorithms to classify binary protein sequences as highly-designable or poorly-designable 
BMC Bioinformatics  2008;9:487.
By using a standard Support Vector Machine (SVM) with a Sequential Minimal Optimization (SMO) method of training, Naïve Bayes and other machine learning algorithms we are able to distinguish between two classes of protein sequences: those folding to highly-designable conformations, or those folding to poorly- or non-designable conformations.
First, we generate all possible compact lattice conformations for the specified shape (a hexagon or a triangle) on the 2D triangular lattice. Then we generate all possible binary hydrophobic/polar (H/P) sequences and by using a specified energy function, thread them through all of these compact conformations. If for a given sequence the lowest energy is obtained for a particular lattice conformation we assume that this sequence folds to that conformation. Highly-designable conformations have many H/P sequences folding to them, while poorly-designable conformations have few or no H/P sequences. We classify sequences as folding to either highly – or poorly-designable conformations. We have randomly selected subsets of the sequences belonging to highly-designable and poorly-designable conformations and used them to train several different standard machine learning algorithms.
By using these machine learning algorithms with ten-fold cross-validation we are able to classify the two classes of sequences with high accuracy – in some cases exceeding 95%.
PMCID: PMC2655094  PMID: 19014713
4.  Shape-dependent designability studies of lattice proteins 
One important problem in computational structural biology is protein designability, that is, why protein sequences are not random strings of amino acids but instead show regular patterns that encode protein structures. Many previous studies that have attempted to solve the problem have relied upon reduced models of proteins. In particular, the 2D square and the 3D cubic lattices together with reduced amino acid alphabet models have been examined extensively and have lead to interesting results that shed some light on evolutionary relationship among proteins. Here we perform designability studies on the 2D square lattice and explore the effects of variable overall shapes on protein designability using a binary hydrophobic-polar (HP) amino acid alphabet. Because we rely on a simple energy function that counts the total number of H-H interactions between non-sequential residues, we restrict our studies to protein shapes that have the same number of residues and also a constant number of non-bonded contacts. We have found that there is a marked difference in the designability between various protein shapes, with some of them accounting for a significantly larger share of the total foldable sequences.
PMCID: PMC2134837  PMID: 18079979

Results 1-4 (4)