Warren J. Gallin

Biological data sets are increasingly becoming information-dense, making it effective to use a computer science-based analysis. We used convolution neural networks (CNN) and the specific CNN architecture Unet to study sponge behavior over time. We analyzed a large time series of hourly high-resolution still images of a marine sponge, Suberites concinnus (Demospongiae, Suberitidae) captured between 2012 and 2015 using the NEPTUNE seafloor cabled observatory, off the west coast of Vancouver Island, Canada. We applied semantic segmentation with the Unet architecture with some modifications, including adapting parts of the architecture to be more applicable to three-channel images (RGB). Some alterations that made this model successful were the use of a dice-loss coefficient, Adam optimizer and a dropout function after each convolutional layer which provided losses, accuracies and dice scores of up to 0.03, 0.98 and 0.97, respectively. The model was tested with five-fold cross-validation. This study is a first step towards analyzing trends in the behavior of a demosponge in an environment that experiences severe seasonal and inter-annual changes in climate. The end objective is to correlate changes in sponge size (activity) over seasons and years with environmental variables collected from the same observatory platform. Our work provides a roadmap for others who seek to cross the interdisciplinary boundaries between biology and computer science.

Voltage-gated potassium ion (Kv) channel proteins respond to changes in membrane potential by changing the probability of K+ flux through an ion-selective pore. Kv channels from different paralogous and orthologous families have widely varying V50 values. The voltage-sensing transmembrane helices (S4) of different channels contain 4-7 basic residues that are responsible for transducing changes in transmembrane potential into the energy required to shift the equilibrium between the open- and closed-channel conformations. These residues also form electrostatic interaction networks with acidic residues in the S2 and S3 helices that stabilize the open and the closed states to different extents. The length and composition of the extracellular loop connecting the S3 and S4 helices (S3-S4 loop) also shape the voltage response. We describe mutagenesis experiments on the jellyfish (Polyorchis penicillatus) Kv1 family jShak1 channel to evaluate how variants of the S3-S4 loop affect the voltage sensitivity of this channel. In combination with changes in the length and composition of the S3-S4 linker we mutated a residue on the S2 helix (N227) that in most Kv1 family channels is glutamate (E226 in mouse Kv1.2, E283 in D. melanogaster Shaker). Some individual loop replacement mutants cause major changes in voltage sensitivity, depending on a combination of length and composition. Pairwise combinations of the loop mutations and the S2 mutations interact to yield quantitatively distinct, non-additive changes in voltage sensitivity. We conclude that the S3-S4 loop interacts energetically with the residue at position N227 during the transitions between open and closed states of the channel.

Voltage-gated potassium (K_v) channels work in concert with other ion channels to determine the frequency and duration of action potentials in excitable cells. Little is known about K_v3 channels from invertebrates, but those that have been characterized generally display slow kinetics. Here, we report the cloning and characterization of jShaw1, the first K_v3 isolated from a cnidarian, the jellyfish Polyorchis penicillatus, in comparison with mouse K_v3.1 and K_v3.2. Using a two-electrode voltage clamp on Xenopus laevis oocytes expressing the channels, we compared steady-state and kinetic properties of macroscopic currents. jShaw1 is fast activating, and opens at potentials approximately 40 mV more hyperpolarized than the mouse K_v3 channels. There is an inverse relationship between the number of positive charges on the voltage sensor and the half-activation voltage of the channel, contrary to what would be expected with the simplest model of voltage sensitivity. jShaw1 has kinetic characteristics that are substantially different from the mammalian K_v3 channels, including a much lower sensitivity of early activation rates to incremental voltage changes, and a much faster voltage-dependent transition in the last stages of opening. jShaw1 opening kinetics were affected little by pre-depolarization voltage, in contrast to both mouse channels. Similar to the mouse channels, jShaw1 was half-blocked by 0.7 mmol l^–1 tetraethyl ammonium and 5 mmol l^–1 4-aminopyridine. Comparison of sequence and functional properties of jShaw1 with the mouse and other reported K_v3 channels helps to illuminate the general relationship between amino acid sequence and electrophysiological activity in this channel family.

The V oltage-gated K⁺C hannel D ata B ase (VKCDB) ( http://vkcdb.biology.ualberta.ca ) makes a comprehensive set of sequence data readily available for phylogenetic and comparative analysis. The current update contains 2063 entries for full-length or nearly full-length unique channel sequences from Bacteria (477), Archaea (18) and Eukaryotes (1568), an increase from 346 solely eukaryotic entries in the original release. In addition to protein sequences for channels, corresponding nucleotide sequences of the open reading frames corresponding to the amino acid sequences are now available and can be extracted in parallel with sets of protein sequences. Channels are categorized into subfamilies by phylogenetic analysis and by using hidden Markov model analyses. Although the raw database contains a number of fragmentary, duplicated, obsolete and non-channel sequences that were collected in early steps of data collection, the web interface will only return entries that have been validated as likely K ⁺ channels. The retrieval function of the web interface allows retrieval of entries that contain a substantial fraction of the core structural elements of VKCs, fragmentary entries, or both. The full database can be downloaded as either a MySQL dump or as an XML dump from the web site. We have now implemented automated updates at quarterly intervals.

Background: Voltage-gated ion channels are membrane proteins containing a selective pore that allows permeable ions to transit the membrane in response to a change in the transmembrane voltage. The typical selectivity filter in potassium channels is formed by a tetrameric arrangement of the carbonyl groups of the conserved amino-acid sequence Gly-Tyr-Gly. This canonical pore is opened or closed by conformational changes that originate in the voltage sensor (S4), a transmembrane helix with a series of positively charged amino acids. This sensor moves through a gating pore formed by elements of the S1, S2 and S3 helices, across the plane of the membrane, without allowing ions to pass through the membrane at that site. Recently, synthetic mutagenesis studies in the Drosophila melanogaster Shaker channel and analysis of human disease-causing mutations in sodium channels have identified amino acid residues that are integral parts of the gating-pore; when these residues are mutated the proteins allow a non-specific cation current, known as the omega current, to pass through the gating-pore with relatively low selectivity. Results: The N.at-K _v 3.2 potassium channel has an unusual weak inward rectifier phenotype. Several mutations of two amino acids in the voltage sensing (S4) transmembrane helix change the phenotype to a typical delayed rectifier. The inward rectifier channels (wild-type and mutant) are sensitive to 4-aminopyridine (4-AP) but not tetra-ethyl ammonium (TEA), whereas the delayed rectifier mutants are sensitive to TEA but not 4-AP. The inward rectifier channels also manifest low cation selectivity. The relative selectivity for different cations is sensitive to specific mutations in the S4 helix, Conclusion: N.at-K _v 3.2, a naturally occurring potassium channel of the Kv3 sequence family, mediates ion permeation through a modified gating pore, not the canonical, highly selective pore typical of potassium channels. This channel has evolved to yield qualitatively different ion permeability when compared to all other members of this gene family.

Background: The accuracy of protein secondary structure prediction has steadily improved over the past 30 years. Now many secondary structure prediction methods routinely achieve an accuracy (Q3) of about 75%. We believe this accuracy could be further improved by including structure (as opposed to sequence) database comparisons as part of the prediction process. Indeed, given the large size of the Protein Data Bank (>35,000 sequences), the probability of a newly identified sequence having a structural homologue is actually quite high.Results: We have developed a method that performs structure-based sequence alignments as part of the secondary structure prediction process. By mapping the structure of a known homologue (sequence ID >25%) onto the query protein's sequence, it is possible to predict at least a portion of that query protein's secondary structure. By integrating this structural alignment approach with conventional (sequence-based) secondary structure methods and then combining it with a "jury-of-experts" system to generate a consensus result, it is possible to attain very high prediction accuracy. Using a sequence-unique test set of 1644 proteins from EVA, this new method achieves an average Q3 score of 81.3%. Extensive testing indicates this is approximately 4–5% better than any other method currently available. Assessments using non sequence-unique test sets (typical of those used in proteome annotation or structural genomics) indicate that this new method can achieve a Q3 score approaching 88%.Conclusion: By using both sequence and structure databases and by exploiting the latest techniques in machine learning it is possible to routinely predict protein secondary structure with an accuracy well above 80%. A program and web server, called PROTEUS, that performs these secondary structure predictions is accessible athttp://wishart.biology.ualberta.ca/proteus. For high throughput or batch sequence analyses, the PROTEUS programs, databases (and server) can be downloaded and run locally.

Background: Studies of the structure-function relationship in proteins for which no 3D structure is available are often based on inspection of multiple sequence alignments. Many functionally important residues of proteins can be identified because they are conserved during evolution. However, residues that vary can also be critically important if their variation is responsible for diversity of protein function and improved phenotypes. If too few sequences are studied, the support for hypotheses on the role of a given residue will be weak, but analysis of large multiple alignments is too complex for simple inspection. When a large body of sequence and functional data are available for a protein family, mature data mining tools, such as machine learning, can be applied to extract information more easily, sensitively and reliably. We have undertaken such an analysis of voltage-gated potassium channels, a transmembrane protein family whose members play indispensable roles in electrically excitable cells. Results: We applied different learning algorithms, combined in various implementations, to obtain a model that predicts the half activation voltage of a voltage-gated potassium channel based on its amino acid sequence. The best result was obtained with a k-nearest neighbor classifier combined with a wrapper algorithm for feature selection, producing a mean absolute error of prediction of 7.0 mV. The predictor was validated by permutation test and evaluation of independent experimental data. Feature selection identified a number of residues that are predicted to be involved in the voltage sensitive conformation changes; these residues are good target candidates for mutagenesis analysis. Conclusion: Machine learning analysis can identify new testable hypotheses about the structure/function relationship in the voltage-gated potassium channel family. This approach should be applicable to any protein family if the number of training examples and the sequence diversity of the training set that are necessary for robust prediction are empirically validated. The predictor and datasets can be found at the VKCDB web site [1].

The family of voltage-gated potassium channels comprises a functionally diverse group of membrane proteins. They help maintain and regulate the potassium ion-based component of the membrane potential and are thus central to many critical physiological processes. VKCDB (V oltage-gated potassium [K] C hannel D ataB ase) is a database of structural and functional data on these channels. It is designed as a resource for research on the molecular basis of voltage-gated potassium channel function. Voltage-gated potassium channel sequences were identified by using BLASTP to search GENBANK and SWISSPROT. Annotations for all voltage-gated potassium channels were selectively parsed and integrated into VKCDB. Electrophysiological and pharmacological data for the channels were collected from published journal articles. Transmembrane domain predictions by TMHMM and PHD are included for each VKCDB entry. Multiple sequence alignments of conserved domains of channels of the four Kv families and the KCNQ family are also included. Currently VKCDB contains 346 channel entries. It can be browsed and searched using a set of functionally relevant categories. Protein sequences can also be searched using a local BLAST engine. VKCDB is a resource for comparative studies of voltage-gated potassium channels. The methods used to construct VKCDB are general; they can be used to create specialized databases for other protein families. VKCDB is accessible at http://vkcdb.biology.ualberta.ca .