# SMILES alignment: a dynamic programming approach for the alignment of metabolites and other small organic molecules

## TL;DR

This paper introduces a new method for aligning small organic molecules using SMILES format, helping to understand structural changes in biochemical pathways.

## Contribution

A novel dynamic programming approach for SMILES-based molecular alignment that quantifies structural transformations in metabolic pathways.

## Key findings

- The algorithm successfully aligns atoms in known chemical transformations, such as those in the Krebs cycle.
- It identified a midpoint dissimilarity peak in cyclical pathways and a progressive decrease in similarity in linear pathways.
- The method provides insights into structural changes during transitions between linear and cyclical biochemical pathways.

## Abstract

There is a need for computational approaches to compare small organic molecules based on chemical similarity or for evaluating biochemical transformations. No tool currently exists to generate global molecular alignments for small organic molecules. The study introduces a new approach to molecular alignment in the Simplified Molecular Input Line Entry System (SMILES) format. This method leverages programming and scoring alignments to minimize differences in electronegativity, here using a measure of atomic partial charges to address the challenge of understanding structural transformations in reaction pathways. This can be applied to study transitions from linear to cyclical pathways.

The proposed method is based on the Needleman-Wunsch algorithm for sequence alignment, but it uses a modified scoring function for different input data. Validation against a benchmarked dataset from the Krebs cycle, based on the known chemical transformations in the pathway, confirmed the efficacy of the approach in aligning atoms that are known to be the same across the transformation. The algorithm also quantified each transformation of metabolites in the Pentose Phosphate Pathway and in Glycolysis. The method was used to study the difference in chemical similarity over transformations between linear and cyclical pathways. The study found a midpoint dissimilarity peak in cyclical pathways (particularly the Krebs Cycle) and a progressive decrease in molecular similarity in linear pathways, consistent with expectations.

The study introduces an algorithm that quantifies molecular transformations in metabolic pathways. The algorithm effectively highlights structural changes and was applied to a hypothesis about the transition from linear to cyclical structures. The software, which provides valuable insights into molecular transformations, is available at: https://github.com/24atang/SMILES-Alignment.git