@article {CaltechAUTHORS_https://authors.library.caltech.edu/id/eprint/99083,
    title ="ASM-Clust: classifying functionally diverse protein families using alignment score matrices",
    author = "Speth, Daan R. and Orphan, Victoria J.",
    
    
    
    
    month = "October",
    year = "2019",
    
    
    
    url = "https://resolver.caltech.edu/CaltechAUTHORS:20191004-102107177",
    note = "The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY 4.0 International license. \n\nbioRxiv preprint first posted online Oct. 3, 2019. \n\nThis work was supported by the US Department of Energy, Office of Science, Office of Biological and Environmental Research under award number DE-SC0016469 to Victoria J. Orphan. Daan R. Speth was supported by the Netherlands Organisation for Scientific Research, Rubicon award 019.153LW.039.",
    revision_no = "10",
    abstract = "Rapid advances in sequencing technology have resulted in the availability of genomes from organisms across the tree of life. Accurately interpreting the function of proteins in these genomes is a major challenge, as annotation transfer based on homology frequently results in misannotation and error propagation. This challenge is especially pressing for organisms whose genomes are directly obtained from environmental samples, as interpretation of their physiology and ecology is often based solely on the genome sequence. For complex protein (super)families containing a large number of sequences, classification can be used to determine whether annotation transfer is appropriate, or whether experimental evidence for function is lacking. Here we present a novel computational approach for de novo classification of large protein (super)families, based on clustering an alignment score matrix obtained by aligning all sequences in the family to a small subset of the data. We evaluate our approach on the enolase family in the Structure Function Linkage Database.",
}
@article {CaltechAUTHORS_https://authors.library.caltech.edu/id/eprint/94306,
    title ="FIND: Identifying Functionally and Structurally Important Features in Protein Sequences with Deep Neural Networks",
    author = "Murali, Ranjani and Hemp, James",
    
    
    
    
    month = "March",
    year = "2019",
    
    
    
    url = "https://resolver.caltech.edu/CaltechAUTHORS:20190401-084306692",
    note = "The copyright holder for this preprint is the author/funder, who has granted bioRxiv a license to display the preprint in perpetuity. It is made available under a CC-BY-NC-ND 4.0 International license. \n\nThis work was funding by DARPA’s CwC program through ARO (W911NF-15-1-0543), the Moore Foundation(Award to VJO: GBMF3780), the Department of Energy (Award to VJO: DE-SC0016469) and the Center for Dark Biosphere Investigations (NSF Award to VJO: OCE-0939564). We are grateful to Daan Speth for reading the manuscript at multiple stages of completion and for his useful comments throughout the process. We would also like to thank Dr. Robert Gennis for reading the manuscript and offering helpful comments, Grayson Chadwick for useful discussion on bioinformatics strategies and Haley Sapers for her instructive critique of the figures. ",
    revision_no = "11",
    abstract = "The ability to correctly predict the functional role of proteins from their amino acid sequences would significantly advance biological studies at the molecular level by improving our ability to understand the biochemical capability of biological organisms from their genomic sequence. Existing methods that are geared towards protein function prediction or annotation mostly use alignment-based approaches and probabilistic models such as Hidden-Markov Models. In this work we introduce a deep learning architecture (Function Identification with Neural Descriptions or FIND) which performs protein annotation from primary sequence. The accuracy of our methods matches state of the art techniques, such as protein classifiers based on Hidden Markov Models. Further, our approach allows for model introspection via a neural attention mechanism, which weights parts of the amino acid sequence proportionally to their relevance for functional assignment. In this way, the attention weights automatically uncover structurally and functionally relevant features of the classified protein and find novel functional motifs in previously uncharacterized proteins. While this model is applicable to any database of proteins, we chose to apply this model to superfamilies of homologous proteins, with the aim of extracting features inherent to divergent protein families within a larger superfamily. This provided insight into the functional diversification of an enzyme superfamily and its adaptation to different physiological contexts. We tested our approach on three families (nitrogenases, cytochrome bd-type oxygen reductases and heme-copper oxygen reductases) and present a detailed analysis of the sequence characteristics identified in previously characterized proteins in the heme-copper oxygen reductase (HCO) superfamily. These are correlated with their catalytic relevance and evolutionary history. FIND was then applied to discover features in previously uncharacterized members of the HCO superfamily, providing insight into their unique sequence features. This modeling approach demonstrates the power of neural networks to recognize patterns in large datasets and can be utilized to discover biochemically and structurally important features in proteins from their amino acid sequences.",
}
@techreport {CaltechAUTHORS_https://authors.library.caltech.edu/id/eprint/86060,
    title ="Methane on Mars and Habitability: Challenges and Responses",
    author = "Yung, Yuk L. and Chen, Pin",
    journal = "Astrobiology",
    volume = "18",
    
    pages = "1221-1242",
    month = "January",
    year = "2018",
    
    issn = "1557-8070",
    
    url = "https://resolver.caltech.edu/CaltechAUTHORS:20180426-110252898",
    note = "Study Report prepared for the Keck Institute for Space Studies (KISS).\n\nWe dedicate this report to JPL scientist Mark Allen, whose spirit of generosity inspired many of the conversations and much of the work that brought us together. We thank the Director of the Keck Institute for Space Studies, Tom Prince, the Executive Director, Michele Judd, and the fantastic KISS staff for hosting our workshop and providing a wonderful environment for these discussions.",
    revision_no = "53",
    abstract = "Recent measurements of methane (CH_4) by the Mars Science Laboratory (MSL) now confront us with robust data that demand interpretation. Thus far, the MSL data have revealed a baseline level of CH_4 (∼0.4 parts per billion by volume [ppbv]), with seasonal variations, as well as greatly enhanced spikes of CH_4 with peak abundances of ∼7\u2009ppbv. What do these CH_4 revelations with drastically different abundances and temporal signatures represent in terms of interior geochemical processes, or is martian CH_4 a biosignature? Discerning how CH_4 generation occurs on Mars may shed light on the potential habitability of Mars. There is no evidence of life on the surface of Mars today, but microbes might reside beneath the surface. In this case, the carbon flux represented by CH_4 would serve as a link between a putative subterranean biosphere on Mars and what we can measure above the surface. Alternatively, CH_4 records modern geochemical activity. Here we ask the fundamental question: how active is Mars, geochemically and/or biologically? In this article, we examine geological, geochemical, and biogeochemical processes related to our overarching question. The martian atmosphere and surface are an overwhelmingly oxidizing environment, and life requires pairing of electron donors and electron acceptors, that is, redox gradients, as an essential source of energy. Therefore, a fundamental and critical question regarding the possibility of life on Mars is, “Where can we find redox gradients as energy sources for life on Mars?” Hence, regardless of the pathway that generates CH_4 on Mars, the presence of CH_4, a reduced species in an oxidant-rich environment, suggests the possibility of redox gradients supporting life and habitability on Mars. Recent missions such as ExoMars Trace Gas Orbiter may provide mapping of the global distribution of CH_4. To discriminate between abiotic and biotic sources of CH_4 on Mars, future studies should use a series of diagnostic geochemical analyses, preferably performed below the ground or at the ground/atmosphere interface, including measurements of CH_4 isotopes, methane/ethane ratios, H_2 gas concentration, and species such as acetic acid. Advances in the fields of Mars exploration and instrumentation will be driven, augmented, and supported by an improved understanding of atmospheric chemistry and dynamics, deep subsurface biogeochemistry, astrobiology, planetary geology, and geophysics. Future Mars exploration programs will have to expand the integration of complementary areas of expertise to generate synergistic and innovative ideas to realize breakthroughs in advancing our understanding of the potential of life and habitable conditions having existed on Mars. In this spirit, we conducted a set of interdisciplinary workshops. From this series has emerged a vision of technological, theoretical, and methodological innovations to explore the martian subsurface and to enhance spatial tracking of key volatiles, such as CH_4.",
}