Recent developments in deep learning, coupled with an increasing number of sequenced proteins, have led to a breakthrough in life science applications, in particular in protein property prediction. There is hope that deep learning can close the gap between the number of sequenced proteins and proteins with known properties based on lab experiments. Language models from the field of natural language processing have gained popularity for protein property predictions and have led to a new computational revolution in biology, where old prediction results are being improved regularly. Such models can learn useful multipurpose representations of proteins from large open repositories of protein sequences and can be used, for instance, to predict protein properties. The field of natural language processing is growing quickly because of developments in a class of models based on a particular model—the Transformer model. We review recent developments and the use of large-scale Transformer models in applications for predicting protein characteristics and how such models can be used to predict, for example, post-translational modifications. We review shortcomings of other deep learning models and explain how the Transformer models have quickly proven to be a very promising way to unravel information hidden in the sequences of amino acids.

Summary The key basic helix–loop–helix (bHLH) transcription factor in iron (Fe) uptake, FER‐LIKE IRON DEFICIENCY‐INDUCED TRANSCRIPTION FACTOR (FIT), is controlled by multiple signaling pathways, important to adjust Fe acquisition to growth and environmental constraints. FIT protein exists in active and inactive protein pools, and phosphorylation of serine Ser272 in the C‐terminus, a regulatory domain of FIT, provides a trigger for FIT activation. Here, we use phospho‐mutant activity assays and study phospho‐mimicking and phospho‐dead mutations of three additional predicted phosphorylation sites, namely at Ser221 and at tyrosines Tyr238 and Tyr278, besides Ser 272. Phospho‐mutations at these sites affect FIT activities in yeast, plant, and mammalian cells. The diverse array of cellular phenotypes is seen at the level of cellular localization, nuclear mobility, homodimerization, and dimerization with the FIT‐activating partner bHLH039, promoter transactivation, and protein stability. Phospho‐mimicking Tyr mutations of FIT disturb fit mutant plant complementation. Taken together, we provide evidence that FIT is activated through Ser and deactivated through Tyr site phosphorylation. We therefore propose that FIT activity is regulated by alternative phosphorylation pathways.

The interaction between plants and plant pathogens can have significant effects on ecosystem performance. For their growth and development, both bionts rely on amino acids. While amino acids are key transport forms of nitrogen and can be directly absorbed from the soil through specific root amino acid transporters, various pathogenic microbes can invade plant tissues to feed on different plant amino acid pools. In parallel, plants may initiate an immune response program to restrict this invasion, employing various amino acid transporters to modify the amino acid pool at the site of pathogen attack. The interaction between pathogens and plants is sophisticated and responses are dynamic. Both avail themselves of multiple tools to increase their chance of survival. In this review, we highlight the role of amino acid transporters during pathogen infection. Having control over the expression of those transporters can be decisive for the fate of both bionts but the underlying mechanism that regulates the expression of amino acid transporters is not understood to date. We provide an overview of the regulation of a variety of amino acid transporters, depending on interaction with biotrophic, hemibiotrophic or necrotrophic pathogens. In addition, we aim to highlight the interplay of different physiological processes on amino acid transporter regulation during pathogen attack and chose the LYSINE HISTIDINE TRANSPORTER1 (LHT1) as an example.