The chemical and physical characteristics of a biomacromolecule influence the final deposition patterns formed during drying a droplet of that biomacromolecule solution on a material’s surface. Deposition patterns are not only complex and characteristic but also highly reproducible. These patterns can thus be used as “fingerprints” for biomacromolecule screening and categorization. 

Most sequence-specific DNA binding proteins recognize and bind to their target DNA sequence with a high affinity by utilizing structural domains that make sequence-specific contacts with the DNA bases in the major groove. These contacts utilize extensive non-covalent bonding and hydrophobic interactions. In contrast, non-specific protein–DNA interactions occur with much lower affinity. The reason for this low binding affinity is that most non-sequence-specific interactions between protein and DNA involve only weak interactions, primarily electrostatic in nature, between the protein and the negatively charged DNA backbone. DNA binding proteins such as transcription factors, constitute about 10 % of the protein-coding genes in eukaryotic genomes and play pivotal roles in the regulation of chromatin structure and gene expression by binding to DNA.

• C2H2-zinc finger (ZF) is the major TF family in eukaryotes. It is involved in several cellular processes, including cancer progression and metastasis formation. Its alteration is involved in neurodegeneration, skin disease, and diabetes. Moreover, it has a Key role in the development and differentiation of several tissues.

In this study, we collected and screened the drying droplet pattern images (on the coated plates) of a library of commercially available DNA sequences. By training the deep learning neural network with two classes of specific and non-specific interactions, it should be possible to distinguish that novel binding candidate.