August 2021 - phenixbb - phenix-online.org

RapiData 2020 applications open
by Smith, Clyde 16 Nov '21

16 Nov '21

Applications are now open for the 2020 edition of RapiData at SSRL. RapiData is a practical course in macromolecular X-ray diffraction data collection, data processing and structure solution. The aim of the RapiData course is to educate and train young scientists in data collection and processing methods at synchrotron beamlines, using state-of-the-art software and instrumentation. The course will be held in sunny California at the Stanford Synchrotron Radiation Lightsource (SSRL) located on the SLAC National Accelerator Laboratory campus in Menlo Park, CA, from March 30 - April 4 2020. Course organizers for 2020 are Silvia Russi, Clyde Smith and Jeney Wierman. The course will comprise hands-on experiments at the SSRL beamlines, software tutorials, and lectures on the following topics: Specimen preparation, Data collection strategies, X-ray light sources, X-ray detectors, Data reduction, Structure solution by MAD, SAD and Molecular Replacement, Complementary methods (spectroscopy and small angle scattering). Please visit http://smb.slac.stanford.edu/news/rapidata/rapidata-2020/ and use the links on the "registration" page to apply for the course. The application deadline is November 10 2019. Successful applicants will be notified shortly thereafter, and invited to register and book accommodation at the SLAC Guesthouse at that time. A limited amount of travel support funding may be available. Please direct questions to Silvia Russi (srussi(a)slac.stanford.edu), Jeney Wierman (jwierman(a)slac.stanford.edu) or Clyde Smith (csmith(a)slac.stanford.edu).

1 2

copy R-free array from reference mtz using command line?
by Aaron Thompson 30 Aug '21

30 Aug '21

PhenixBB, I know this can be done via GUI, but is it possible to copy the R-free array from a reference mtz to a new mtz using the command line? I don’t see this as an option in phenix.reflection_file_converter. Thank you, Aaron

3 2

NORA workshop on AlphaFold2 and RoseTTAFold, 31 August and 1 September
by Randy John Read 27 Aug '21

27 Aug '21

Hi, I’ve just learned that this online workshop is generally open, not just to members of the Norwegian Artificial Intelligence Consortium (NORA), and it’s free! Information, including the current programme and a link to registration, can be found here: https://www.nora.ai/events/AlphaFoldv2.0-and-RoseTTAFold-workshop%20.html I’ll be talking a bit about experiences with models from AlphaFold2 and RoseTTAFold, but the real draw should be talks by people who really understand these two deep learning algorithms: Minkyung Baek (first author on the RoseTTAFold paper, from David Baker’s group) and a representative of the DeepMind team that developed the two versions of AlphaFold. Sameer Velankar will also be there to talk about the exciting new AlphaFold database at the EBI. Best wishes, Randy Read ----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Fax: +44 1223 336827 Hills Road E-mail: rjr27(a)cam.ac.uk Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk

2 2

Re: [phenixbb] NORA workshop on AlphaFold2 and RoseTTAFold, 31 August and 1 September
by Randy John Read 27 Aug '21

27 Aug '21

9 am CEST (Norway). Pretty early for our North American colleagues, especially those on the west coast! > On 27 Aug 2021, at 05:18, Robert Rose <rbrose2(a)ncsu.edu> wrote: > > Does the workshop start on August 31 at 9 am GMT? > > On Thu, Aug 26, 2021 at 2:27 PM Randy John Read <rjr27(a)cam.ac.uk> wrote: > Hi, > > I’ve just learned that this online workshop is generally open, not just to members of the Norwegian Artificial Intelligence Consortium (NORA), and it’s free! Information, including the current programme and a link to registration, can be found here: > > https://www.nora.ai/events/AlphaFoldv2.0-and-RoseTTAFold-workshop%20.html > > I’ll be talking a bit about experiences with models from AlphaFold2 and RoseTTAFold, but the real draw should be talks by people who really understand these two deep learning algorithms: Minkyung Baek (first author on the RoseTTAFold paper, from David Baker’s group) and a representative of the DeepMind team that developed the two versions of AlphaFold. Sameer Velankar will also be there to talk about the exciting new AlphaFold database at the EBI. > > Best wishes, > > Randy Read > > ----- > Randy J. Read > Department of Haematology, University of Cambridge > Cambridge Institute for Medical Research Tel: +44 1223 336500 > The Keith Peters Building Fax: +44 1223 336827 > Hills Road E-mail: rjr27(a)cam.ac.uk > Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk > > > _______________________________________________ > phenixbb mailing list > phenixbb(a)phenix-online.org > http://phenix-online.org/mailman/listinfo/phenixbb > Unsubscribe: phenixbb-leave(a)phenix-online.org ----- Randy J. Read Department of Haematology, University of Cambridge Cambridge Institute for Medical Research Tel: +44 1223 336500 The Keith Peters Building Fax: +44 1223 336827 Hills Road E-mail: rjr27(a)cam.ac.uk Cambridge CB2 0XY, U.K. www-structmed.cimr.cam.ac.uk

1 0

Peptide bond to symmetry related molecule
by Damian Ekiert 24 Aug '21

24 Aug '21

Hi All, We are refining several structures in which the N-terminal residue of chain A appears to form a peptide bond to the C-terminal residue of its symmetry mate, chain A. In reality, the target protein consists of 4 identical helical repeats that pack end to end. Slippage and averaging over the 4 possible registers leads to the appearance of an infinitely long polymer in the crystal, with no breaks between protomers. Is it possible to enforce a peptide bond to symmetry mates in Phenix, and what is the easiest way to do this? I have found a several cases in the archives on how to deal with symmetry clashes, but struggling to figure out how to deal with a peptide bond. Thanks for your help! Best, Damian - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - Damian Ekiert Assistant Professor Skirball Institute | NYU School of Medicine Depts. Cell Biology and Microbiology damian.ekiert(a)ekiertlab.org www.bhabhaekiertlab.org

3 4

Postdoctoral position in Computational Biochemistry
by Yasser Almeida 23 Aug '21

23 Aug '21

Dear all, The group of Computational Biochemistry of Prof. Elsa Sánchez-García at the University of Duisburg-Essen is announcing a Postdoc position focused on the computational design of chemical tools and inhibitors for the study of proteins and protein-protein interactions involved in cell state transition. The deadline for applications has been changed to September 30th. More information can be found at https://www.uni-due.de/computational-biochemistry. Best regards, Yasser -- Dr. Yasser Almeida-Hernandez Postdoctoral Researcher Computational Biochemistry group T01 R03 D45 Faculty of Biology University of Duisburg-Essen Universitätsstr. 2, 45117 Essen Email: yasser.almeida-hernandez(a)uni-due.de <%20yasser.almeida-hernandez(a)uni-due.de> Phone: +49 201 183 2457

1 0

Project Scientist and Postdoctoral research positions in XFEL, ML, and material science computational methods
by Aaron Brewster 18 Aug '21

18 Aug '21

Hello, I'm excited to announce that two positions are immediately open in my group in the fields of material science, serial X-ray crystallography, and machine learning. Links to postings: Computational Project Scientist - https://lbl.taleo.net/careersection/2/jobdetail.ftl?job=93507 Computational Postdoctoral Scholar - https://lbl.taleo.net/careersection/2/jobdetail.ftl?job=93506 In these exciting roles, you will work in the Molecular Biophysics and Integrated Bioimaging Division on a project that will research adding advanced spatial and symmetry aware neural network training methods to small-molecule chemical crystallography and material science data processing. This will use AI methods developed as part of a close collaboration between teams at Lawrence Berkeley National Labs (LBNL), University of Connecticut, and MIT. The larger group is researching crystal engineering of hybrid coordination polymers towards achieving rational control over the optoelectronic properties of low-dimensional nanostructures. These positions are focused on the chemical crystallography aspect of the program, but the successful candidates will be expected to participate in all aspects of the program, including data collection at X-ray free electron lasers worldwide, data analysis on supercomputing facilities, structure solution, research on AI driven structure prediction, paper writing, and software management. Both roles are 3 year appointments. This work will be part of the Department of Energy Integrated Computational and Data Infrastructure for Science <https://insidehpc.com/2021/08/doe-funds-15m-for-integrated-computational-an…> Research program. Specific details follow. Berkeley Lab is committed to Inclusion, Diversity, Equity and Accountability (IDEA) <https://diversity.lbl.gov/ideaberkeleylab/> and strives to continue building community with these shared values and commitments. Berkeley Lab is an Equal Opportunity and Affirmative Action Employer. We heartily welcome applications from women, minorities, veterans, and all who would contribute to the Lab’s mission of leading scientific discovery, inclusion, and professionalism. In support of our diverse global community, all qualified applicants will be considered for employment without regard to race, color, religion, sex, sexual orientation, gender identity, national origin, disability, age, or protected veteran status. *Computational Project Scientist* *What You Will Do:* - Design and train neural networks that allow determination of the unit cell and crystal orientations of small molecule crystals from sparse, randomly oriented diffraction patterns. This will involve curating training sets from simulated datasets and from existing and new experimental data that will be collected. - Research methods to scale training and use of these neural networks on multiple nodes in high performance computing environments. - Lead data collection and processing at X-ray free electron sources. - Maintain a detailed record of work performed. - Document and manage code developed using standard source control methods. - Publish in peer-reviewed journals. - Present research at seminars and conferences. - Assist Principal Investigator in developing grant proposals. - Document and maintain existing and new data processing software, including the cctbx build system. *What is Required:* - 2-5 years of relevant experience beyond the highest degree in Physics, Statistics, Crystallography, Artificial Intelligence, or a related field. - A strong background in Python, C++, or an equivalent platform for scientific computing. - A track record of critical thinking, problem solving, and mathematical intuition, demonstrated through publications and presentations. *Desired Qualifications:* - Familiarity with general Python scientific data packages such as NumPy, SciPy, and pandas. - Versed in modern coding practices for scientific computing, including algorithm design and evaluation, source code management under a revision control system like git, and incorporation of readable documentation and comments. - Expertise in computational approaches in at least one of the areas related to the research is required, including but not limited to small-molecular crystallography, structure solution and refinement, unit cell determination, neural network design, training, and evaluation, high speed computing, image analysis, large dataset analysis, graph theory, error propagation and statistics, signal processing, and GPU acceleration. - Familiarity with small-molecule material design and synthesis. - Strong analytical and organizational skills. - Excellent interpersonal, oral, and written communication skills. - Ability to work independently as well as part of a diverse team. *Computational **Postdoctoral Scholar* *What You Will Do:* - Design and train neural networks to model diffraction from small molecule crystals. This will involve curating training sets from simulated datasets and from existing and new experimental data that will be collected. - Assist in data collection and processing at X-ray free electron sources. - Maintain a detailed record of work performed. - Document and manage code developed using standard source control methods. - Publish in peer-reviewed journals. - Present research at seminars and conferences. * What is Required:* - PhD degree in Physics, Statistics, Crystallography, Artificial Intelligence, or a related field. - Experience in computational approaches in at least one of the areas related to the research, including but not limited to small-molecular crystallography, structure solution and refinement, unit cell determination, neural network design, training, and evaluation, high speed computing, image analysis, large dataset analysis, graph theory, error propagation and statistics, signal processing, and GPU acceleration. - Strong analytical and organizational skills. - Excellent interpersonal, oral, and written communication skills. - Ability to work independently as well as part of a diverse team. * Desired Qualifications:* - Familiarity with Python, C++, and general scientific data packages such as NumPy, SciPy, and pandas. - Familiarity with small-molecule material design and synthesis. Please do not hesitate to reach out to me for more information. Thank you, Aaron S. Brewster, PhD Computational Research Scientist, Molecular Biophysics and Integrated Bioimaging Division Lawrence Berkeley National Laboratory 1 Cyclotron Rd., Bldg. 33R0345 Berkeley, CA 94720

1 0

Specifying import-time dependencies in _ext libraries?
by Russell M. Taylor II 18 Aug '21

18 Aug '21

I've developed an mmtbx_probe_ext module that makes use of objects defined in iotbx_pdb. I've added that library to LIBS= in the SConscript file, and it is able to link and run properly. However, when a Python script imports mmtbx_probe_ext, that load fails unless the same script has previously imported from iotbx: The following works: from iotbx.map_model_manager import map_model_manager import mmtbx_probe_ext as probe But the following fails: import mmtbx_probe_ext as probe from iotbx.map_model_manager import map_model_manager ImportError: /home/taylorr/rlab/cctbx-reduce-python27/build/lib/mmtbx_probe_ext.so: undefined symbol: _ZNK5iotbx3pdb9hierarchy5chain5atomsEi How do I tell Sconscript (or what do I do in my _bpl file) to tell Python that it needs to load the other module first? Thanks, Russ Taylor --- Russell M. Taylor II, Ph.D. russ(a)reliasolve.com taylor(a)redbudlabs.com rtaylor(a)aqueti.com russ(a)outloveracism.org www.cs.unc.edu/~taylorr

1 0

Postdoc position in San Antonio, TX
by Yogesh Gupta 17 Aug '21

17 Aug '21

Dear Colleagues: My laboratory at the University of Texas Health Science Center at San Antonio is seeking an enthusiastic Postdoctoral Fellow to work on an NIH-funded project on SARS-CoV-2 and RNA modification. We will use X-ray crystallography, cryo-EM, and cell-based methods to address basic mechanistic questions and develop novel antivirals. These two papers describe our most recent work on SARS-CoV-2. Nature Communications. 2021 Jun 2; 12: 3287. PMID: 34078893. DOI: 10.1038/s41467-021-23594-y <https://www.nature.com/articles/s41467-021-23594-y> Nature Communications. 2020 Jul 24;11(1):3718. PMID: 32709886. DOI: 10.1038/s41467-020-17496-8 <https://www.nature.com/articles/s41467-020-17496-8> Preference will be given to applicants with demonstrated experience in purification and structural characterization of protein-nucleic acid complexes by X-ray crystallography with a strong interest in learning cryo-EM. An ideal candidate should hold (or soon expect to hold) a Ph.D. or equivalent degree in biochemistry/biophysics/structural biology and have no more than 3 years of postdoc experience. Salary is competitive and commensurate with experience. San Antonio is the nation’s seventh-largest city and is located at the edge of the beautiful Texas Hill County. San Antonio offers a rich, multi-cultural community, affordable cost of living, excellent weather, and a thriving biomedical industry. There is no state income tax in Texas. All Postdoctoral appointments are designated as security-sensitive positions. The University of Texas Health Science Center at San Antonio is an Equal Opportunity/Affirmative Action Employer, including protected veterans and persons with disabilities. If interested in applying, please send a short description of your research accomplishments and a CV with contact information of 3 references to Dr. Yogesh Gupta (email: guptay(a)uthscsa.edu). Yogesh -- Yogesh Gupta, PhD Assistant Professor of Biochemistry & Structural Biology Greehey Children's Cancer Research Institute University of Texas Health Science Center 8403 Floyd Curl Drive, San Antonio, TX, USA 78229 http://ccri.uthscsa.edu/YGupta.html

1 0

Using AutoBuild to Complete a Model
by Whitley, Matthew J 17 Aug '21

17 Aug '21

Hello all, I have a very large protein with 16 copies in the asymmetric unit (nearly 15,000 total residues). Upon initial model building, ~30 residues at the N-terminus of each identical chain were not built. After building the reliable parts of the model and filling in other gaps and then refining that partial (but nearly complete) model, the N-terminal density has now improved to the degree that I can manually build a number of the missing residues in Coot. However, that means manually building the same series of residues over and over for each of the 16 identical chains. Is there a way to get AutoBuild to add these N-terminal residues for me to spare me the tedium of repetitive manual building? What I’ve gleaned from the documentation so far: >From the phenix.fit_loops documentation, it seems like that tool only works for filling in chain breaks, i.e. there is already modeled protein at each end of the missing region. That’s not the case for me because I’m trying to work backward toward the N-terminus. If I am reading the phenix.autobuild documention correctly, it sounds like there are two modes: (i) Standard mode - rebuild the whole protein from scratch, which does give the possibility of adding my N-terminal residues. However, I don’t want my nearly complete existing model to be rebuilt, I just want AutoBuild to try to add the missing N-terminal sequence to the existing model. (ii) Rebuild-in-place, which rebuilds the current model but doesn’t add or subtract any atoms. So that sounds like it’s not what I’m looking for because it won’t add the N-terminal sequence AND it rebuilds the existing model, which I want to keep exactly as is. Is there some series of options in AutoBuild that will allow me to just add missing residues (provided in the sequence file) but keep all the other currently existing residue in the input model exactly as they are? Happy to hear any advice you can give. Thanks, Matthew --- Matthew J. Whitley, Ph.D. Research Instructor Department of Pharmacology & Chemical Biology University of Pittsburgh School of Medicine

3 3