Pattern recognition on random trees associated to protein functionality families

In this paper, we address the problem of identifying protein functionality using the information contained in its aminoacid sequence. We propose a method to define sequence similarity relationships that can be used as input for classification and clustering via well known metric based statistical methods. In our examples, we specifically address two problems of supervised and unsupervised learning in structural genomics via simple metric based techniques on the space of trees 1)Unsupervised detection of functionality families via K means clustering in the space of trees, 2)Classification of new proteins into known families via k nearest neighbour trees. We found evidence that the similarity measure induced by our approach concentrates information for discrimination. Classification has the same high performance than others VLMC approaches. Clustering is a harder task, though, but our approach for clustering is alignment free and automatic, and may lead to many interesting variations by choosing other clustering or classification procedures that are based on pre-computed similarity information, as the ones that performs clustering using flow simulation, see (Yona et al 2000, Enright et al, 2003).

💡 Research Summary

The manuscript tackles the long‑standing problem of protein function prediction using only the primary amino‑acid sequence, but departs from the conventional alignment‑based paradigm. Instead, each protein sequence is modeled as a Variable‑Length Markov Chain (VLMC). The VLMC is estimated via the Probabilistic Suffix Tree (PST) algorithm, which yields a rooted context tree whose nodes correspond to variable‑length past symbols (contexts) and whose edges carry transition probabilities.

To compare proteins, the authors adopt the Balding‑Ferguson‑Friedman‑Smyth (BFFS) distance, a metric defined on the space of rooted trees. For two trees t and y, the distance is d(t,y)=∑_{v∈V} φ(v)