In molecular biology, Prof Zhang’s lab developed a genomic language model-enhanced algorithm that discovered novel classes of noncanonical circular transcripts from RNA sequencing data. They have characterised the genomic distributions and sequence features of such novel circular RNAs and demonstrated their biological significance and clinical relevance as biomarkers for lung cancer, paving the way for translational research on circular RNAs.
Prof Zhang’s early work showed, for the first time, that RNA Polymerase II transcribes microRNAs in humans, mice and rice. His team has then extended the method to identify stress-responsive microRNAs that respond to environmental stresses such as salt and drought in rice and model plants like Arabidopsis.
In the machine learning area, using the Traveling Salesman Problem (TSP) as a model, Prof Zhang revealed easy-to-difficult phase transition patterns for combinatorial optimisation problems, including the TSP and maximum Boolean satisfaction problems. Furthermore, he identified the representation precision, e.g., the number of bits used to measure the distances between cities in the TSP, as the control parameter or determining factor of the phase transitions in optimisation problems such as the TSP and many scheduling problems.
ORCID ID
SCOPUS ID