MIT Researchers Analyze Protein Similarity Across Species

“Low complexity regions” (LCRs), also called compositionally biased regions, are regions in protein sequences that contain repeats of an amino acid or its motifs. While LCRs are well known to be involved in various processes in eukaryotic cells, much remains undiscovered regarding specific features and functions. That’s why researchers at MIT developed a new technique that allows them to analyze similarities and differences in these sequences in greater detail.

The researchers, whose work was published in eLife, focused on a broad view of LCRs, rather than looking at specific sequences within one individual or species. Individually studying these sequences can be time-consuming, so the team utilized bioinformatics to evaluate multiple LCRs simultaneously. They found that, while LCRs can significantly vary between proteins and species, they often share similar roles, such as helping proteins join larger-scale assemblies in the nucleolus. 

“Instead of looking at specific LCRs and their functions, which might seem separate because they’re involved in different processes, our broader approach allows us to see similarities between their properties, suggesting that maybe the functions of LCRs aren’t so disparate after all,” says Byron Lee, lead author and MIT graduate student.

The team utilized a dotplot matrix to visually compare the amino acid sequences and generate specific images of each protein. Then, they developed a computational pipeline to uncover more about the spatial dynamics and relationships of LCRs. This algorithm allowed them to compare thousands of these matrices at the same time and categorize LCRs based on which amino acids were most frequently repeated.

“What we wanted to do is take a step back and, instead of looking at individual LCRs, to try to take a look at all of them and to see if we could observe some patterns on a larger scale that might help us figure out what the ones that have assigned functions are doing, and also help us learn a bit about what the ones that don’t have assigned functions are doing,” says Jaberi-Lashkari.

The team compared proteins across eight different species and found that some LCRs were highly conserved, but that wasn’t the case for all. Some LCRs identified in plants had specific proteins used to scaffold their cell walls, which are not observed in other types of organisms. However, they also identified many similar LCRs between species, like those involved in constructing the extracellular matrix and upholding the nucleolus.

“These sequences seem to be important for the assembly of certain parts of the nucleolus,” Lee says. “Some of the principles that are known to be important for higher order assembly seem to be at play because the copy number, which might control how many interactions a protein can make, is important for the protein to integrate into that compartment.”

The team plans to expand their LCR analysis to include additional species for future work. “There’s so much to explore because we can expand this map to essentially any species,” Lee says. “That gives us the opportunity and the framework to identify new biological assemblies.”