Deep neural networks have been successfully applied to diverse fields of computer vision. However, they only outperform human capacities in a few cases. The ability of deep neural networks versus human experts to classify microscopy images was tested on biofilm colonization patterns formed on sulfide minerals composed of up to three different bioleaching bacterial species attached to chalcopyrite sample particles. A low number of microscopy images per category (<600) was sufficient for highly efficient computational analysis of the biofilm's bacterial composition. The use of deep neural networks reached an accuracy of classification of ∼90% compared to ∼50% for human experts. Deep neural networks outperform human experts' capacity in characterizing bacterial biofilm composition involved in the degradation of chalcopyrite. This approach provides an alternative to standard, time-consuming biochemical methods.

Leptospirillum ferriphilum plays a major role in acidic, metal-rich environments, where it represents one of the most prevalent iron oxidizers. These milieus include acid rock and mine drainage as well as biomining operations. Despite its perceived importance, no complete genome sequence of the type strain of this model species is available, limiting the possibilities to investigate the strategies and adaptations that Leptospirillum ferriphilum DSM 14647 ^T (here referred to as Leptospirillum ferriphilum ^T ) applies to survive and compete in its niche. This study presents a complete, circular genome of Leptospirillum ferriphilum ^T obtained by PacBio single-molecule real-time (SMRT) long-read sequencing for use as a high-quality reference. Analysis of the functionally annotated genome, mRNA transcripts, and protein concentrations revealed a previously undiscovered nitrogenase cluster for atmospheric nitrogen fixation and elucidated metabolic systems taking part in energy conservation, carbon fixation, pH homeostasis, heavy metal tolerance, the oxidative stress response, chemotaxis and motility, quorum sensing, and biofilm formation. Additionally, mRNA transcript counts and protein concentrations were compared between cells grown in continuous culture using ferrous iron as the substrate and those grown in bioleaching cultures containing chalcopyrite (CuFeS ₂ ). Adaptations of Leptospirillum ferriphilum ^T to growth on chalcopyrite included the possibly enhanced production of reducing power, reduced carbon dioxide fixation, as well as elevated levels of RNA transcripts and proteins involved in heavy metal resistance, with special emphasis on copper efflux systems. Finally, the expression and translation of genes responsible for chemotaxis and motility were enhanced. IMPORTANCE Leptospirillum ferriphilum is one of the most important iron oxidizers in the context of acidic and metal-rich environments during moderately thermophilic biomining. A high-quality circular genome of Leptospirillum ferriphilum ^T coupled with functional omics data provides new insights into its metabolic properties, such as the novel identification of genes for atmospheric nitrogen fixation, and represents an essential step for further accurate proteomic and transcriptomic investigation of this acidophile model species in the future. Additionally, light is shed on adaptation strategies of Leptospirillum ferriphilum ^T for growth on the copper mineral chalcopyrite. These data can be applied to deepen our understanding and optimization of bioleaching and biooxidation, techniques that present sustainable and environmentally friendly alternatives to many traditional methods for metal extraction.

Reactive oxygen species (ROS) cause oxidative stress and growth inhibition by inactivation of essential enzymes, DNA and lipid damage in microbial cells. Acid mine drainage (AMD) ecosystems are characterized by low pH values, enhanced levels of metal ions and low species abundance. Furthermore, metal sulfides, such as pyrite and chalcopyrite, generate extracellular ROS upon exposure to acidic water. Consequently, oxidative stress management is especially important in acidophilic leaching microorganisms present in industrial biomining operations, especially when forming biofilms on metal sulfides. Several adaptive mechanisms have been described, but the molecular repertoire of responses upon exposure to pyrite and the presence of ROS are not thoroughly understood in acidophiles. In this study the impact of the addition of H₂O₂ on iron oxidation activity in Acidithiobacillus ferrooxidans DSM 14882^T was investigated. Iron(II)- or sulfur-grown cells showed a higher sensitivity toward H₂O₂ than pyrite-grown ones. In order to elucidate which molecular responses may be involved, we used shot-gun proteomics and compared proteomes of cells grown with iron(II)-ions against biofilm cells, grown for 5 days in presence of pyrite as sole energy source. In total 1157 proteins were identified. 213 and 207 ones were found to have increased levels in iron(II) ion-grown or pyrite-biofilm cells, respectively. Proteins associated with inorganic sulfur compound (ISC) oxidation were among the latter. In total, 80 proteins involved in ROS degradation, thiol redox regulation, macromolecule repair mechanisms, biosynthesis of antioxidants, as well as metal and oxygen homeostasis were found. 42 of these proteins had no significant changes in abundance, while 30 proteins had increased levels in pyrite-biofilm cells. New insights in ROS mitigation strategies, such as importance of globins for oxygen homeostasis and prevention of unspecific reactions of free oxygen that generate ROS are presented for A. ferrooxidans biofilm cells. Furthermore, proteomic analyses provide insights in adaptations of carbon fixation and oxidative phosphorylation pathways under these two growth conditions.