Details: These are reads from high-throughput sequencing of crosslinked RNA pulled down with an anti-Argonaute antibody (CLIP-seq).

Data:
Available via GEO

Format: These files are in single line FASTA format. The header line contains a 6nt degenerate barcode associated with each sequenced read. The same degenerate barcode is reproduced in the first 6nt positions of the read sequence line.

Example:

Random barcode: >CTAGAA
AGO bound read: CTAGAACCGTAGCCCTGGCGGATGCCTGGTAGGTGGAAGCGT 

Details: These are sequences and details of Argonaute (AGO) binding sites/peaks. These peaks have been processed with the following steps:
1) Normalized read counts from all WT and KO libraries are convolved with a Gaussian.
2) Gradient descent/ascent is used to identify "edges" of peaks.
3) Required >50% of WT or KO replicate libraries present in each peak.
4) Sequences and statistics are generated from the region spanning the peaks (binding site).

Data:
CD4_miR155_clipbase.zip

Format: This file is in tab-delimited format. Each line contains the sequence and details for one peak. Multiple attributes/details are present for each peak.

Example:

 [Line: 6296] 
Transcript ID: NM_001205043.1 
Gene Symbol: Jarid2 
Peak Number: 2 
WT read count: 7.02049 
KO read count: 2.44848 
Diff. (WT-KO): 4.57201 
Diff. P-value: 2.57589e-13 
Replicate stat: 12 
Peak start pos: 127 
Peak end pos: 162 
Peak sequence: TAGAGAACTGATTTTGTTTTAGCATTAAACTGTTCAAGTTTTTGTACG 

Details: These are 3' UTR sequences curated from RefSeq and Ensembl (Aug 15 2011). These sequences have been processed with the following steps:
1) Retrieved all 3' UTR definitions from RefSeq and Ensembl.
2) Selected the longest 3' UTR sequence based on gene name.

Data:
mouse.utr.fa.zip

Format: This file is in tab-delimited format. Each line contains one 3' UTR sequence for a gene. The following attributes are present for each gene: Transcript ID, Gene Symbol, Species ID and UTR Sequence.

Example:

 [Line 10682] 
Transcript ID: NM_001205043.1 
Gene Symbol: Jarid2 
Taxonomy ID: 10090 
3' UTR seq.: AAGATGCCG... 

Details: Currently we are converting the entire pipeline into C and an open source R library. Please email aakhan@cbio.mskcc.org for any technical questions. Algorithmic details can be found in supplementary methods of corresponding publications.