Skip to main content
Premium Trial:

Request an Annual Quote

This Week in Nucleic Acids Research: Jan 22, 2014

A pair of researchers from Princeton University present a method for finding functionally important cancer genes based on protein interactions with nucleic acids, peptides, ions, and other small molecules. The duo developed a computational pipeline called CanBind that makes predictions about genes that play a functional part in cancer following an analysis of existing cancer genome data that found more frequent than usual glitches affecting protein residues that participate in such interactions. The CanBind method brings together exome sequence data from individuals' cancers with protein structure information, they say, highlighting proteins that are prone to mutations affecting their binding sites.

A University of Pennsylvania team introduces a statistical strategy for quantifying the expression of specific transcript isoforms from RNA sequencing data. Their approach, dubbed PennSeq, takes into account read distribution across the transcriptome and models instances of non-uniform distribution for each isoform. After using PennSeq to profile isoform-specific expression with RNA-seq data, the researchers show that it could also come up with accurate expression information for transcript isoforms in real RNA-seq datasets generate on Illumina instruments. "Our results indicate superior performance of PennSeq over existing methods," the study authors say, "particularly for isoforms demonstrating sever non-uniformity."

Phylogenetic modeling may offer insights into contributions that duplications make to cancer progression, according to another Nucleic Acids Research study. Researchers from the US and China came up with a phylogenetic model for assessing duplications and deletions in sequence data representing tumor and normal samples from five individuals with different stages of stomach cancer. The analysis revealed elevated duplication and deletion rates in the tumor genomes compared to their counterparts from normal tissue, the team notes, and made it possible to unearth nine genes that are particularly prone to duplication in stomach cancer.