|
|
|
Name of Submitter:
|
Simon Kasif
|
|
Title of proposed idea:
|
Computational Bridges to Experiments (COMBREX)
|
|
What is the major obstacle/challenge in the biomedical research field? What is needed to overcome this obstacle/challenge?
|
|
High throughput sequencing led to discovery of millions of genes whose function is not well understood.
Although the development of high-throughput technologies for answering gene function questions is an important goal of genomic research, at present many important experimental questions related to gene function and phenotype are answered using low throughput experimental methods. How does the scientific community collectively makes broad progress on which genes and gene families and their functional elucidation?
As a proof of concept, we propose to make a substantial progress on closing the gap in gene function understanding and provide an initial annotation for genes without functional annotation (orphan genes), primarily focusing on biochemistry and basic role in biological processes. This can be done in the human genome or selected microbial pathogens. |
|
What emerging scientific opportunity is ripe for investment by the Common Fund?
|
|
Our proposal generally lies in producing better infrastructures and bridges between biologists and computational scientists by building novel infrastructures to solve high priority biology or medical questions by coordinated consortiums that integrate crowd-sourcing, predictive systems, computation, and biological experiments. |
|
What are the potential Common Fund investments that could accelerate scientific progress in this field?
|
|
As a key concept, to involve the largest possible subset of the scientific community in producing functional annotation to high priority genes, COMBREX has established a novel micro-granting mechanism to support experimental validations of gene functions. COMBREX is recruiting experimentalists from the biochemical and microbiological communities through professional organizations such as the ASM. Experimentalists can search the database for gene function predictions in their areas of expertise, and submit brief proposals to experimentally validate those predictions. Proposals go through a rapid peer review; grants are typically on the order of a few thousands to a few tens of thousands of dollars to validate predictions for a small number of genes. As of this writing, multiple experimental bids have been funded to validate over 100 genes and multiple new bids have been submitted. |
|
If a Common Fund program on this topic achieved its objectives, what would be the impact?
|
|
COMBREX is implementing an automated recommendation system to help experimentalists identify sets of proteins to include in an experiment on a given function. Recommendations take into account a number of factors, but a key consideration is impact of the experiment or information gain; i.e., the expected impact of a validation on the annotation of homologous or functionally linked proteins. Information gain can take different forms depending on the method or model used for gene function prediction.
Other criteria include medically important gene phenotypes such as essentiality, pathogenicity, antibiotic resistance, biofilm formation, and growth.
The overall impact of COMBREX type project would be maximal acceleration on closing gene function knowledge gaps. |
|