CAMEL logo Computational and Analytical Molecular Evolution Lab at CARB Nexplorer
People Publications Software Opportunities Links Internal

Welcome

Welcome to the web home of the Computational and Analytical Molecular Evolution Laboratory (Stoltzfus group) at CARB. Use the navigation table at the top of this page to find what you want.

What's hot (latest news from the group)

Mutation: In the half-century since Watson & Crick proposed their "tautomer" model, its been widely assumed that spontaneous errors of replication are the main source of germline mutations. However, in a forthcoming article in Mutation Research, Arlin shows that the context-dependence of nucleotide mutation in mammals looks surprisingly like the context-dependence of oxidative damage due to hydroxyl radical.

The emperor has no codons: in a paper that just appeared, Antezana and Jordan pull the rug out from under 25 years of adaptationist story-telling by showing that the patterns typically referred to as codon preferences and assigned an adaptive cause relating to translation (improving "efficiency" or reducing error) more properly represent frameless motif preferences that, since they recur in non-coding frames, cannot be assigned an adaptive cause relating to translation. The Antezana and Jordan paper is a difficult read: if you have difficulty understanding the randomization method that underlies most of the results, please take a look at this figure.

Evolutionary informatics: In May, the evoinfo working group (supported by NESCent, based on the proposal by Rutger Vos and Arlin Stoltzfus) just had its third meeting focused on interoperability. We heard progress reports on nexml (now supported by DAMBE and Mesquite; pledged support in PAUP* and GARLI) and CDAO (Comparative Data Analysis Ontology).

Synthesis: !Viva la Revolución! At the 2008 Mol Evol Gordon Conference, Arlin enjoyed the weather in Ventura and tried to reclaim the lost Revolutionary Heritage of Evo-Devo. The Revolution Will Not be Televised (nor Published it seems).

Research Areas

Mutation and evolutionary genetics

Contrary to the Modern Synthesis view, biases in mutation (e.g., see bubble plot of mutation rates at right) bias the course of evolution, even adaptive evolution (see Rokyta, et al.), in a predictable manner. This causal influence raises many questions:
  • What is the population-genetic mechanism (see Yampolsky and Stoltzfus, 2001)?
  • How did neo-Darwinists overlook this mechanism, claiming instead that internal causes of direction are impossible (Stoltfus, 2006a)?
  • Can we tease apart mutational effects from fitness effects? (Yampolsky and Stoltzfus, 2005)
  • What kinds of mutation biases exist and what is their impact?
  • Can they cause long-term patterns or trends (Stoltzfus, 2006b)?
  • How often are mutational effects wrongly attributed to "function"?
  • What about developmental biases in phenotypic variation (Stoltzfus, 2006a)?
Our research in this vital and exciting area includes both theoretical modeling and data analysis. Ongoing work addresses the alleged adaptedness of the genetic code (Stoltzfus and Yampolsky, accepted), and the challenges for genomics (Stoltzfus, in prep).

The evolution of introns and intron-containing genes

How did our genomes come to contain hundreds of thousands of introns? As shown by Stoltzfus, et al. (1997), when comparing intron-containing genes from diverse eukaryotic species, the diversity in intron sites is not mainly due to inheritance and loss of primordial introns or to "intron sliding", but to the addition of introns to genes during eukaryotic evolution. Thus the original "introns-early" view in which all or most introns date back to a primordial ancestor cannot be correct. However, a coherent alternative theory that accounts for non-random patterns in the distribution of introns (e.g., phase bias) has yet to be established. A major step forward came in 2004, when Qiu, et al. (Stoltzfus group) and Sverdlov, et al. (Koonin group at NCBI) provided rigorous evidence for Dibb's "protosplice" hypothesis that introns are gained with nucleotide sequence preferences (see the logo at left). Our ongoing research uses sequence analysis, database tools, and computer modelling to evaluate the causes for this pattern, and its consequences for the distribution of introns with respect to reading frame phase and with respect to protein structural features. Recent work includes a demonstration that sequence-based models largely explain non-uniformity in the positions of introns relative to protein structure (De Kee, Gopalan and Stoltzfus, MBE), and a demonstration, using the tubulin gene family, of the inadequacy of simple methods of character reconstruction to account for intron data (Gopalan and Stoltzfus, in prep)

Bioinformatics Tools for Molecular Evolution Research

Development of software tools is an ongoing process in our research group. Most such tools relate to a specific project and have a short lifespan. However, a long-term goal is to develop a database system to automate the evolutionary analysis of gene family data. This system has two main parts: a software pipeline that extracts and collates sets of gene family data extracted from sequence databases, and a database system that provides for phylogenetic inference, querying and analysis. Some of the fruits of these efforts are Bio::NEXUS, our NEXUS applications programming interface in Perl, described by Hladish, et al., 2007, the Nexplorer server interface described by Gopalan, et al., and the thousands of sequence family data sets that can be accessed via Nexplorer.

As of 2006, we are doing some of this work in conjunction with the Evolutionary Informatics Working Group supported by the National Evolutionary Synthesis Center (aka "NESCent"). This working group, headed by Stoltzfus and Rutger Vos (of Bio::Phylo fame), has a mandate to improve interoperability in comparative analysis.

Screenshot showing the Nexplorer interface with intron data for a subset of the ATP Synthase Subunit C family, with control panel (above), datamatrix (right) and gene phylogeny (left), from Fig. 1 of Gopalan, et al., 2006
Gene scrambling in Ciliates (figure from Stoltzfus, 1999) the Flying Spaghetti Monster

And other topics, some related

  • More realistic models of protein evolution
  • Evolution on rough fitness landscapes
  • Integrating mutation-and-altered-development into evolutionary theory
  • RNA editing, gene scrambling, and other strange things
  • Duplicate gene evolution
  • Flying-Spaghetti-Monster (FSM) Theory

Directions to CARB

Institutional Affiliations

Contact information