attach 1_sow.docx

Attachment-1 Attachment-1

STATEMENT OF WORK

Computational discovery for molecular and epigenetic processes

Background:

NIH’s mission is to seek fundamental knowledge about the nature and behavior of living systems and the application of that knowledge to enhance health, lengthen life, and reduce illness and disability. To improve the health of all humans through advances in genomics research. NHGRI is a leading authority in the field of genomics, the mission is to accelerate scientific and medical breakthroughs that improve human health. NHGRI does this by driving cutting-edge research, developing new technologies, and studying the impact of genomics on society. Translational and Functional Genomics Branch, Genomic Functional Analysis Section studies how genome structure and function contribute to health by exploring the genetic, epigenetic and metagenomic basis of human disorders.

A contract is sought for computer science and biomedical research services. The services include bioinformatics research of epigenomic pattern discovery, an important area of genomics relevant to (1) understanding the biology of genomes, (2) understanding the biology of disease, and (3) advancing the practice of personalized medicine. The research activities will be performed in collaboration with Dr. Laura Elnitski’s group at NHGRI, which has made significant accomplishments in the area of discovering epigenetic patterns and biomarkers in the cancer methylome. The project will involve data mining in tumor data sets that include DNA methylation, SNPs, chromosomal interactions, drug sensitivity and CRIPSR editing, among others. The Elnitski lab has published several papers referring to the CpG island methylator phenotype and has proposed that aberrant methylation in each type of cancer follows a regulatory process that can be modeled and tested. The causal events have not been identified for most cancers and the use of machine learning analyses and experimental validation is necessary to sort through enormous amounts of data.

Purpose and Objectives:

The purpose of this procurement is to identify causal mutations, altered expression, copy number variants and other events like micro-environment that are responsible for aberrant methylation, in order to identify candidate genes and develop therapeutic applications in cancers of all types. The project requires computer science expertise to apply new machine learning algorithms, to find high frequency events and distinguish events in affected vs unaffected tumors. Valid experimental design of the computational studies as well as wet lab experiments must be generated through this written application for successful review. First year goals should be fully explained to generate all data within the timeline specified. Genetic algorithms and CRIPSR targets should be included. Deliverables will include project design, prioritized testable hypotheses, and fully prepared manuscripts for high impact, peer-reviewed computational biology journals.

As the project is highly likely to influence design of bioinformatic and computational approaches in the future, applicants should demonstrate prominent roles in professional societies, student training and curriculum development at the graduate and post-doctoral levels. Additionally, international leadership experience will be expected in order to facilitate implementation of novel approaches into University level coursework. Evidence of advanced doctoral degree teaching experience, and appointments in multiple disciplinary areas are essential, including, medical, translational, molecular biology, computer science, and engineering. Manuscript production and patent applications are essential components of the project. Applicants must demonstrate prior evidence as PI of competitive awards, grants and patent applications. Applicants must also demonstrate prior publications in the area of focus of this contract.

Specific tasks to be accomplished:

Describe in detail the application of machine learning approaches to the datasets from large consortium data or other datasets, determine sensitivity and specificity of each analysis, define discrete targets for follow-up by experimental testing. Datasets for validation testing should be identified and explained. Computational approaches should be enumerated not only for their individual attributes, but also for compound learning outcomes that could be achieved. Fundamental to this process is direct mentorship of one or more matriculated graduate student trainees in computational and experimental graduate programs, whose trainees from underrepresented demographic categories are strongly encouraged, in order to provide a mentoring and teaching experience to the student. The student(s) shall be included on the published manuscript.

Contractor Requirement:

The Contractor shall complete the following tasks:

1. The Contractor shall define datasets to be used. Enumerate and implement steps taken to weight features, determine optimal numbers of relevant features, and establish how to evaluate optimal outcomes for prioritizing causal events in aberrant methylation. Estimated timeline: 16 weeks.

2. The Contractor shall define additional cohorts of data and implement validation studies. Enumerate all steps for validation and datasets to be used. Repeat analysis after parameter tuning. Estimated timeline: 8 weeks.

3. The Contractor shall describe in detail steps to apply additional machine learning approaches to datasets. Implement the approaches and cross validate results. Justify methodology to be used. Calculate classification performance with respect to distinct patterns of DNA methylation. Adjust parameters and explain performance attributes. Compare results among all orthologous approaches. Proposals aiming to perform deep learning without indication of relevant features and what they contribute to the results will not be considered. Estimated timeline: 8 weeks.

4. The Contractor shall describe and implement categorization of additional features of data: mutational load, DNA repair, mutational signatures, fusion genes, methylation targets, among others. Propose additional features that should be examined. Summarize how all features are to be integrated and evaluated. Estimated timeline: 4 weeks.

5. The Contractor shall describe all approaches that are likely to be included to plot data using graphical visualization and assessment of statistical parameters. Calculate significance of combinations of mutations focused on particular subsets of samples, and assess significance over the entire collection of samples. Interpret data for biological processes that are common, and unique, across different tumor types. Estimated timeline: 4 weeks.

6. The Contractor shall prioritize mutational pathways that are implicated as causal events, illustrate how prioritization would be implemented.

7. The Contractor shall develop working models of mechanisms that can be tested using a variety of methods in the wet lab. Discuss options. Estimated timeline: 4 weeks.

8. The Contractor shall describe analysis approaches and analyze data generated in wet lab procedures, 16 weeks.

9. The Contractor shall draft manuscript. Timeline: 4 weeks from the time the data are complete.

10. The Contractor shall revise manuscript, address missing analyses, edit text and figures. Timeline: 4 weeks.

11. The Contractor shall respond to reviewer comments. Revise and finalize manuscript, add additional experiments suggested by reviewers, resubmit. Timeline: 4 weeks.

Deliverables:

1. The Contractor shall address deliverables in each time interval.

2. The Contractor shall prepare and present progress reports including primary data to be delivered at the end of each week of the project, including power point presentations and other methods of data visualization.

3. The Contractor shall adapt and adjust to new deliverables, which may be required at the end of each meeting, depending on the materials presented and conclusions drawn.

4. The Contractor shall check Timelines for individual components are mandatory but are not sequential and may overlap.

5. The Contractor shall draft manuscript which will shall be submitted within 4 weeks of the date the data are complete suitable for submission to a high impact, biologically focused journal.

6. The Contractor shall provide revised manuscript submitted within 4 weeks of receiving feedback.

7. The Contractor shall provide a final manuscript within 4 weeks of receiving feedback, and no later than the end of the base year.

Reporting requirements:

· Each week a progress report shall be made via conference or videocast interface along with electronic copies such as power point and other data illustrations. The Government will give feedback and assess the biological findings and suggest next steps in the project.

Anticipated Period of Performance:

This will be awarded as a base year with four options:

Base: August 23, 2020 to August 22, 2021 Option 1: August 23, 2021 to August 22, 2022 Option 2: August 23, 2022 to August 22, 2023 Option 3: August 23, 2023 to August 22, 2024 Option 4: August 23, 2024 to August 22, 2025

Tasks for Option Year(s) 1-5: After completion of the research for the Base Year, the tasks to be performed in the option year(s) will consist of expanding the work to address new experimental analyses and updating the original approach, incorporating the new data, expanding to new cancer types, implementing new computational advances. Describe timelines for and specific aims of future work anticipated for each annual renewal.

Contract Type:

Firm-Fixed Price Place of Contract Performance:

Work will be provided at the Contractor’s site Additional Clauses::

Data Rights:

· FAR Clause 52.227-14, Rights in Data - General (May 2014)

· FAR Clause 52.227-17, Rights in Data--Special Works (December 2007)

· FAR Clause 52.217-8, Option to Extend Services;

· FAR Clause 52.217-9, Option to Extend The Term of the Contract

Special Considerations:

Not Applicable

Government Furnished Equipment:

The government will not be providing any equipment.