Identifying A-site locations on ribosome-protected mRNA fragments using Integer Programming
Identifying the A-site and P-site locations on ribosome-protected mRNA fragments from Ribo-Seq experiments is a fundamental step in the quantitative analysis of transcriptome-wide translation properties at the codon level. Ahmed, N. et al developed an integer programming method to identify the A-site location by maximizing an objective function that reflects the fact that the ribosome’s A-site on ribosome-protected fragments must reside between the second and stop codons of an mRNA. This identifies the A-site location as a function of the fragment’s size and the reading frame in Ribo-Seq data generated from S. cerevisiae, E. coli and mouse embryonic stem cells. The correctness of the identified A-site locations is demonstrated by showing that this method, as compared to others, yields the largest ribosome density at established stalling sites. By providing greater accuracy and utilization of a wider range of fragment sizes, this method increases the signal-to-noise ratio of underlying biological signals associated with translation elongation at the codon length scale [1].
RiboA outputs three sets of A-site density profiles: (1) the A-site reads per nucleotide, (2) the A-site reads per nucleotide mapped to Frame 0 by applying the transformation that for reads in frame 1 and 2 the offset is reduced by 1 and 2, respectively, (3) the A-site reads per codon.
RiboA assumes the ribosomes are undergoing steady-state translation, and it can only be applied to steady-state ribosome profiling data. RiboA is not appropriate for datasets from non-steady-state experiments, such as the ribosome run-off experiments where initiation is blocked by antibiotics, such as harringtonine treatment.
The key script to identify A-site offsets and to generate A-site density profiles is here. Specifically, the function "run_offset" is the entry point to identify A-site offsets, and the function "run_profile" is the entry point to generate A-site density profiles. Computationally literate users can run this script in command line or adapt the script to their own need.
E-mail: epo2@psu.edu
[1] Ahmed, N., Sormanni, P., Ciryam, P. et al. Identifying A- and P-site locations on ribosome-protected mRNA fragments using Integer Programming. Sci Rep 9, 6256 (2019). https://doi.org/10.1038/s41598-019-42348-x