Griffith Lab

The Griffith lab is a combined research group driven by the interests of twin scientists Malachi Griffith (Malachi's website) and Obi Griffith (Obi's website). The focus of the lab is on developing methods of applied bioinformatics for personalized medicine and improved cancer care.

Our research is committed to development of open access and open source resources for cancer genome analysis. Research projects cover a wide spectrum of cancer informatics and clinical statistics with an emphasis on translation and application. Specifically, we use computational methods for the analysis of large cancer datasets at the molecular level (DNA, RNA and protein) to identify markers for diagnosis, prognosis and drug response prediction in cancer. We have contributed to the early development of methods for analysis of transcriptional regulation (ORegAnno) and RNA-seq analysis and visualization (Alexa Platform).

The group is engaged in a large number of tumor sequencing projects for AML, breast, liver, lung, and other cancers, investigating primary, relapse and drug resistant tumors. To this end we have worked with others at the McDonnell Genome Institute to develop end-to-end pipelines for clinical cancer sequencing that automate state-of-the-art methods for sequence alignment, somatic variation detection, RNA sequence analysis, and the integration of these data types into user-friendly reports of the most clinically relevant genome and transcriptome changes in a tumor or cohort of tumors (Genome Modeling System). To aid in this effort our group has developed databases, knowledgebases, and web tools for interrogating the druggable genome (DGIDB), driver mutations (DoCM), and interpretations of clinically actionable variants in cancer (CIViC). The group is also actively involved in the identification and scoring of tumor neoantigens and development of related software for design of human cancer vaccines (pVac-Seq).

In addition to our basic and clinical research interests, we are also passionate about the scholarship of teaching and learning. We have made substantial contributions to the training and education of tomorrow's bioinformaticians through our involvement in CBW and CSHL workshops and the BioStars forum. We are currently developing a bioinformatics and clinical informatics training program that takes a practical, hands-on approach to cancer genome analysis for personalized medicine.

We are currently seeking highly talented and motivated graduate students and post-docs with skills that span the spectrum from molecular biologist to software/web developer. For more information, please see Careers.

Research in our lab focuses on applying genomic and information science technologies to personalized medicine in cancer. Projects tend to involve (1) development of new methods, software tools, databases or web resources for the analysis and interpretation of genomic data; or (2) comprehensive genomic/informatics analysis of n-of-1 or tumor cohorts using existing open-source bioinformatic tools (aligners, variant callers, etc.).

Trainees interested in joining the lab may find the following suggestions/resources useful:

  • Use Biostars.org. It contains many posts with suggestions of books, websites, and courses focused on bioinformatics. For example, this post lists web-based resources for learning bioinformatics: BioStars.
  • Learn R. For those focused on the analysis/translation end of bioinformatics, this is the single most powerful tool.
  • Learn at least one scripting language (Perl or Python). Code Academy has a good Python intro.
  • Learn some Ruby/Rails and/or Javascript. We are increasingly creating web resources such as dgidb.org, docm.info and civicdb.org.
  • Take at least some formal computer science training (one or two introductory computer science courses).
  • Learn at least basics of good software development practices: for example, software version control (e.g., github).
  • Develop key skills such as statistics (classic stats and monte carlo methods), machine learning, data wrangling, and other big data techniques.
  • Have at least some hands-on familiarity with wet lab molecular biology (sequencing protocols, tissue culture, PCR, etc.).