Sternberg Lab Homepage  /  Projects

Current Projects

Gene function and networks.

We try to comprehend gene function and the significance of variants in a number of ways. We are part of a C. elegans knock-out by knock-in consortium. We are targeting C. elegansgenes with human disease relevance, conserved genes of unknown function, and nematode-specific genes that might be involved in parasitism. We are also exploring the use of transcriptomes as phenotype. We find they are highly sensitive, and can identify phenotypes in what experts have long thought were ‘wild type' animals. We are assessing the functional consequences of missense variants from human genetic studies using C. elegans. Our initial project focused on autism spectrum disorder associated variants. We find that more than half of the ASD variant that we can study have effects on gene function in C. elegansorthologs. We are now focusing on specific genes for which we can comprehend with respect to the dauer decision discussed above.

Cell migration.

We are studying aspects of cell migration, focusing on the male linker cell because it undergoes a striking and stereotyped long-range migration that involves distinct phases and directions. Our single-cell transcriptional profiling of the linker cell at distinct stages identified many classes of interesting genes. One class comprises acetylcholine receptors, both ionotropic and metaotropic, which we are studying in collaboration with Mihoko Kato (Pomona College). We have discovered that GAR-3, a muscarinic receptor, and Gq are involved in efficient linker cell migration. We also found a novel transmembrane protein LINKIN, that is necessary for attachment of the linker cell to the trailing vas deferens cells as it migrates. With collaborator Tsui-Fen Chou (Caltech), we are studying the roles of interacting proteins in LINKIN function. For example, our proteomic studies indicated that RUVB proteins bind the LINKIN intracellular domain and knocking down RUVB-1 or RUVB-2 lead to the same linker cell adhesion defect as does knockdown of LINKIN.

Neural circuits and computation projects.

C. elegans has a numerically simple nervous system. Its entire connectome has been described, 40 years ago for the 302 neuron hermaphrodite. How can we not understand it yet? Part of the problem is lack of tools. However, C. elegansresearchers, including our collaborators Aravi Samuel, Vlad Susoy (Harvard University and Vivek Venkatachalam (Northeastern University), can image the entire nervous system in behaving animals. Moreover, our lab developed an efficient system (cGAL) to make each neuron genetically accessible so that the full set of optogenetic and molecular genetic techniques can be applied systematically. We are making a complete set of transcriptional Drivers that allow expression in single neuron types in the hermaphrodite and male. Another advance in using genetics for subtle phenotypes is our ability to make clean loss-of-function mutants using CRISPR technology and also true revertants that provide isogenic controls(our "STOP-IN" method). While these tools can be improved, they are a great advance over five years ago. The final piece is our being able to "think like a worm," and understanding what the worm's brain computes. We are thus pursuing two projects designed to confront and prevail over the complexity of an animal's nervous system. In one project, we are analyzing how the worm computes on environmental inputs and chooses between rapid reproduction or diapause (the dauer larval state). This decision takes place over about 12 hours and involves many sensory neurons and interneurons. We are computationally modeling aspects of the decision-making process and the specific circuits with collaborator Cengiz Pehlevan (Harvard University). The second project addresses how the male senses his mating partner, and whether he has a neural representation of her location. Taking advantage of the now known male connectome, whole male tail imaging, and cGAL, we are developing computational models with collaborator Scott Linderman (Stanford University) as we elucidate the neural circuits involved in each "step" of male copulatory behavior building on our comprehensive cell ablation studies of male-specific neurons.

Nematode-specific chemical communication.

Our long-term collaboration with Frank Schroeder (Cornell University) has identified hundreds of related nematode-specific small molecules, many of which we found to have biological effects primarily in social communication. While we are studying the effects of some as regulators of dauer decision (see above), we want to understand their regulation. Regulatory studies are better done by studying the enzymes that control levels of small molecules rather than mass spec assays; hence, we are trying to identify the enzymes that control the production and degradation of specific ascarosides. One approach is to knockout each candidate enzyme with our STOP-IN method and profile the mutant metabolomes. If we see an interesting signal in the metabolomic profile, we test the isogenic control, and also examine developmental, physiological and behavioral phenotypes.

Organizing biological information.

We have a number of projects that seek to organize biological information to enhance research speed and allow comparative genomic studies. We are involved in WormBase, which organizes information about C. elegansand other nematodes (collaboration with Lincoln Stein at the Ontario Institute for Cancer Research and Kevin Howe at the European Bioinformatics Institute); the Gene Ontology Consortium, which organizes information about gene product function in all organisms using the GO ontologies (Collaobration with Paul Thomas (USC), Mike Cherry (Stanford), Judy Blake (Jackson Labs) and Chris Mungall (LBNL); the Alliance of Genome Resources, which is an umbrella organization that is harmonizing information for the GO, WormBase, FlyBase, mouse, rat, budding yeast and zebrafish. As part of these efforts we developed a single sentence level search that allows efficient text-mining (Textpresso), a gene function predictor, and artificial intelligence applications that summarize information in ontology graphs (SoBA) and writes human readable text summaries of gene function (concise descriptions for the Alliance). Our experience with these information resources made us realize that we need to make scientific publication knowledge-base friendly – with authors describing their experiments with controlled vocabularies – and to capture the ~50% of experimental results that are not now reported, we founded, which publishes single experiment, peer-reviewed articles. By having complete control of the publication process, we are in a position to experiment with some aspects of scholarly communication.