fastMSA: Accelerating Multiple Sequence Alignment with Dense Retrieval on Protein Language

Evolutionarily related sequences provide information for the protein structure and function. Multiple sequence alignment, which includes homolog searching from large databases and sequence alignment, is efficient to dig out the information and assist protein structure and function prediction, whose efficiency has been proved by AlphaFold. Despite the existing tools for multiple sequence alignment, searching homologs from the entire UniProt is still time-consuming. Considering the success of AlphaFold, foreseeably, large- scale multiple sequence alignments against massive databases will be a trend in the field. It is very desirable to accelerate this step. Here, we propose a novel method, fastMSA, to improve the speed significantly. Our idea is orthogonal to all the previous accelerating methods. Taking advantage of the protein language model based on BERT, we propose a novel dual encoder architecture that can embed the protein sequences into a low-dimension space and filter the unrelated sequences efficiently before running BLAST. Extensive experimental results suggest that we can recall most of the homologs with a 34-fold speed-up. Moreover, our method is compatible with the downstream tasks, such as structure prediction using AlphaFold. Using multiple sequence alignments generated from our method, we have little performance compromise on the protein structure prediction with much less running time. fastMSA will effectively assist protein sequence, structure, and function analysis based on homologs and multiple sequence alignment.

PREDICTION OF FUNCTIONAL, STRUCTURAL AND STABILITY CHANGES IN PMM2 GENE ASSOCIATED WITH NEPHROTIC SYNDROME USING COMPUTATIONAL ANALYSIS

Objective: Nephrotic syndrome defines as a disorder with a group of symptoms like proteinuria, hypoalbuminemia, hyperlipidemia, and edema. PMM2 encodes phosphomannosemutase protein enzyme involved in the synthesis of N-glycan. Methods: Different Insilico analysis tools: SIFT, PolyPhen, PROVEAN, SNPandGO, MetaSNP, PhDSNP, MutPred, I-Mutant, STRUM, PROCHECK-Ramachandran, COACH and ConSurf, were used to check the effect of nsSNP on protein structure and function. Results: The genetic polymorphism in the PMM2 gene was retrieved from NCBI ClinVar and UniProtKB. Total 20 SNPs were predicted most significant and responsible for disease-causing and decrease protein stability. Conclusion: This study helps to discover disease-causing deleterious SNPs with different computational tools and gives information about potent SNPs.

GlyConnect: a glycan-based conjugation extension of the GlycoDelete technology

Recently, our lab developed GlycoDelete, a technology suite that allows a radical simplification of eukaryotic N-glycosylation. The technology allows to produce glycoproteins that carry single GlcNAc, LacNAc, or LacNAc-Sia type glycans on their N-linked glycosylation sequons. GlycoDelete-type N-glycans are uniquely suited for glycan-based conjugation purposes, as these provide a short, homogeneous and hydrophilic link to the protein backbone. Targeting GlycoDelete-glycans allows for highly site-specific conjugation at sites in the protein which are normally occupied by bulky glycans, thus ensuring minimal interference with protein structure and function. The current manuscript describes the evaluation and optimization of both chemical and chemo-enzymatic conjugation of molecules onto the GlycoDelete-type glycans of a limited set of benchmark proteins