1 / 7

ENCODE enhancers

ENCODE enhancers. 12/13/2013 Yao Fu Gerstein lab. ‘Supervised’ enhancer prediction. Use peaks as examples to learn chromatin features of binding active regions . Identifying Potential Enhancer-like Elements from Discriminative Model . Human genome in 100bp bins. DNase I. . Positive:

ocean
Download Presentation

ENCODE enhancers

An Image/Link below is provided (as is) to download presentation Download Policy: Content on the Website is provided to you AS IS for your information and personal use and may not be sold / licensed / shared on other websites without getting consent from its author. Content is provided to you AS IS for your information and personal use only. Download presentation by click this link. While downloading, if for some reason you are not able to download a presentation, the publisher may have deleted the file from their server. During download, if you can't get a presentation, the file might be deleted by the publisher.

E N D

Presentation Transcript

  1. ENCODE enhancers 12/13/2013 Yao Fu Gerstein lab

  2. ‘Supervised’ enhancer prediction Use peaks as examples to learn chromatin features of binding active regions • Identifying Potential Enhancer-like Elements from Discriminative Model Human genome in 100bp bins DNase I ... Positive: Overlapping with TF peaks Positive examples ... FAIRE ... Negative examples ... H3K4me3 ... ... Machine learning Features Prediction ... Gencode genes ... Filtering Predicted genes ... Strong H3K4me1 & H3K27ac signal ... Get enhancer list away to genes Yip et al., Genome Biology (2012)

  3. “Unsupervised” Segway/chromHMM • Enhancer “states” from unsupervised segmentations (Hoffman et al. & Ernst et al.) PF TSS E WE T CTCF R enhancer / weak enhancer ENCODE combined segmentation from Segway and chromHMM

  4. Combine with Segway/chromHMM • ~130K enhancer-like elements from Yip et al. • ~ 291K “enhancer state” elements from segmentation. • Intersection : ~71K • http://info.gersteinlab.org/Encode-enhancers

  5. Associate enhancers with target genes • Idea: Histone modifications to predict gene expression. • Another direction is to use whole-genome DNA long-range interaction data • Form distal regulatory networks (~20k distal edges in ENCODE rollout; we extend to edges with ~17k genes) HM signals Expression levels H3K4me1 H3K27ac ... Gene 1 Gene 2 Gene 3 ... Scale GM12878 Strong H1-hESC Cell lines HeLa-S3 Hep-G2 K562 Weak ... Enhancer 2. Find TFs binding enhancers in cell lines with strong HMs Gene 3. Draw distal edges from TFs to targets 1. Find correlated enhancer-target pairs TF Yip et al., Genome Biology (2012)

  6. Cell-line specific enhancers Enhancer

  7. ENCODE Work Products (beyond standardized DCC pipelines) Examples of element lists: • Enhancers • DNAse HS sites • broad vs cell-type tissue specific sites • TF targets • proximal and distal • regulatory networks • Allelic genes (ASE) and TF binding sites (ASB) • Fusion (chimeric) transcripts • Non-coding transcription (contigs or transcripts) • classification (e.g. eRNAs) • High-occupancy (HOT) regions • Regions of active chromatin • Chromatin states (e.g. segmentation) • TF motifs (PWMs & sites)

More Related