AutoGrow Activity
AutoGrow 3.1 is an evolutionary algorithm that optimizes candidate lig
Brought to you by:
jdurrant
Activity for AutoGrow
-
jdurrant posted a comment on discussion AutoGrow Support
Hi Patrick. Much thanks for your interest in AutoGrow, and please forgive my delay in getting back to you. We're currently putting together a new version of AutoGrow with some major improvements. My student Jake Spiegel has rewritten and modernized almost the entire codebase. We hope to submit the work for publication in a few months. I'm not sure why exactly the old version isn't processing your input molecules properly, but I'm confident the new version will be able to handle them if they are amenable...
-
patrick connolly modified a comment on discussion AutoGrow Support
Hi jd/AutoGrow support, I'd like to start out by saying I like the tool, it works pretty well, however I have encountered a strange problem. I have been trying to replicate the first procedure outlined in the paper. In this, you used a panel of all possible permutations of brominated benzene as starting compounds. After a ton of trial and error, I think I at least know the source of the problem. My process for PDB generation is i) download the SD file from Pubchem and ii) convert it to PDB with openbabel....
-
Hi jd/AutoGrow support, I'd like to start out by saying I like the tool, it works pretty well, however I have encountered a strange problem. I have been trying to replicate the first procedure outlined in the paper. In this, you used a panel of all possible permutations of brominated benzene as starting compounds. After a ton of trial and error, I think I at least know the source of the problem. My process for PDB generation is i) download the SD file from Pubchem and ii) convert it to PDB with openbabel....
-
Hi Andrew. Please forgive my ridiculous delay in getting back to you. I wanted to let you know that we have begun to work on Autogrow 4. We intend to make it more modular so that custom filters can be easily added. I expect we won't be ready to publish for some months, but I wanted to let you know we're working on it! All the best.
-
AutoGrow released /AutoGrow_latest/latest_version.txt
-
-
Andrew Hogan posted a comment on discussion AutoGrow Support
Hi there, I was wondering if in the future you had plans for developing a filter for (Gram negative) bacterial cells? Its easy to find in vitro enzyme inhibitors, but very few compounds are capable of crossing the bacterial cell envelope, especially the Gram-negative cell envelope. My lab has an antibiotic development project that I've been working on for a while now. I've been using autogrow successfully for a few weeks, but my supervisor isn't as happy with the results as I expected. My supervisor...
-
Hi Francesco. Did you end up getting the run to work?
-
Hi Francesco. If I'm not mistaken, autogrow can filter my molecular weight. That's one of Lipinski's rules.
-
Francesco Pietra posted a comment on discussion AutoGrow Support
I now noticed that those infos about deleting ligands are contained in log.txt, and where present also in previous autogrow sessions with the two families of initial ligands. I forgot to add the input: -filename_of_receptor 5000_less_TA1_Mg.pdb \ -center_x 325.522 -center_y 448.689 -center_z 344.046 \ -size_x 88 -size_y 94 -size_z 126 \ -additional_autoclickchem_parameters "all_reactions" \ -allow_modification_without_frag_addition TRUE \ -directory_of_source_compounds ./starting_sarco/ \ -directory_of_fragments...
-
I have started a new autogrow session with half of previous starting structures, i.e., those of one more homogeneous family only of structures. On ctrl-C, killing the execution and then restarting with same settings, I noticed a series of infos like thais one: Deleting ligand /home/francesco/work_autogrow/sarco_tubulin/autogrow_sarco_tub_out/generation1/157752.mutant.pdb because it contains only 10 marked ("core") atoms. I should inform that I wanted to preserve the core structure of this family...
-
Hi Jacob: The "dimeric" crossovers are indicated in /support by "number.crossover.pdb.can" As a side chain of the involved starting structure ends with a monosaccharide unit, propagation ("dimerization") occurs there at one free -OH. That is, the genetic algorythm carries out its job in the absence of a filter as to the MW. Is that possible to modify the .py code by inserting an upper limit to the MW? Or does it exist in the literature a Lipinski-like filter for high MW? thanks
-
Hi Jacob: All crossover that are generated are elaborated additions of two molecules from the starting list of potential ligands, i.e., MW ca twice as great as that of the initial potential ligands. I expected that the crossover algorythm is such that two initial potential ligands (actually they are true effective ligands) are superimposed for their common part, than each one is added of other (possibly small) fragments, or merely functionalized (I left this option open too). Is anything wrong in...
-
Hi Jacob: At the new trial, most "Making crossover" are successful. In some cases the above WARNING comes out, and it is not always of the largest structures. we will see what comes out. thanks.
-
No, I had removed the filters. I'll try autogrow repeatedly, thanks. As a matter of fact, I remember to have seen autogrow going beyond the superposition with these molecules. But I wanted to change some settings (I don't remember the settings) and destroyed that job. It seems in line with your suggestion. Hopefully.
-
Autogrow does have settings to optionally eliminate compounds if they violate Lipinski's rules, and one of those rules is MW < 500. Could that be the problem?
-
Hi Francesco. I completely understand the need to avoid posting the molecules publicly! It might help with debugging if I had a copy, though. But before we talk further about that, have you tried running autogrow several times? Evolutionary algorithms depend on a lot of random choices. Sometimes the algorithm gets stuck in an evolutionary dead end, but if you rerun it will take a different, more productive path. Could you try running it several times, and if that still doesn't work we could talk...
-
Hi Jacob I wonder whether autogrow3 is programmed for ligands of such high MW
-
Hi Jacob: a short series of (natural) molecules having the same core and differing in side chains. Both strongly anticancer in vitro and the binding site (common to all) is known from inhibition assays. High molecular weight (>500) and therefore I removed filters to that. Clinically useful molecules are in this range of MW. I want to preserve the core structure (spherical) common to the whole series. Privately I can show you details. I did not modify the side chain in silico to have a longer series...
-
No worries, Francesco! Happy you were able to resolve the question. All the best.
-
Hi Francesco. What do your input molecules look like? Is it just one molecule, or is it multiple ones?
-
WARNING: the execution time of the subroutine identifying the largest common substructure has exceeded the maximum run time provided in the command line; a set of smaller substructures is used for the superposition. With my latest settings, the code sticks at this problem. Relevant settings: -allow_modification_without_frag_addition TRUE \ -number_of_mutants_first_generation 10 -number_of_crossovers_first_generation 10 \ -number_of_mutants 10 -number_of_crossovers 10 \ -top_ones_to_advance_to_next_generation...
-
Sorry for my misview. The different models are not merely rotation of a single model.
-
I noticed that in the example at the tutorial of autogrow3...1.1 the ligand (naphtalene) is presented at different orientations in about one plane. I was reading the opinion of the VINA developer: From: autodock-bounces at scripps.edu [autodock-bounces at scripps.edu] on behalf of Oleg Trott [trott at scripps.edu] Sent: Monday, April 14, 2014 10:44 AM To: autodock at scripps.edu Subject: Re: ADL: Vina: docking results dependent on ligand input conformation? Hi The probability distribution of the...
-
Hi Jacob: As a Debian user since the early 90, I am really surprised by the poor python offerED. Anaconda2 installation was a matter of 1/2 hr, against the >6hr of the debian packages, and your tutorial is running on a vintage vaio desktop (10 years aged). Have a nice Sunday francesco
-
Hi Jacob: the same error (future feature print_function is not defined) comes out whether or not all those packages are (system wide) installed besides phyton-scipy and python-numpy. I had already tried that. On the other hand, by merely definyning a temporary path $PYTHONPATH and then commanding python and, from within python, import scipy gives that error with both debian 8 (stable) and debian 9 (testing). No issues with import numpy. I find it difficult to believe that stable debian has a buggy...
-
Hi Francesco. These are all SciPy errors, so you'll want to get that running. Modifying the autogrow source code (by commenting out critical lines, for example) is likely to cause you problems in the future. It is true that python modules are directories, but it's often best just to install them globally so you can access them from anywhere. That way you don't have to look into the package contents themselves. SciPy installation on Debian is described here: https://www.scipy.org/install.html#ubuntu-debian...
-
Back from walking, I resumed OS debian 9 stretch amd64 in the hope that the print_function future import is in python2.7 there. The error in running the autogrow script was now: File "/usr/lib/python2.7/dist-packages/scipy/optimize/optimize.py", line 769 with warnings.catch_warnings(): ^ SyntaxError: invalid syntax
-
Dear Jacob All above problems arose from the testing (scratch) version of amd64 that I had installed in order to have version 18 of scipy. Then I went back to stable (jessie), with scipy 14, and most was recognized by launching autogrow (after a lot of work in setting whas not recognized and removing that from "import", like for optomize and multiprocessing). The latest attempt at launching autogrow: francesco@tya64:~/softw/AUTOGROW/tutorial$ ./executable_mod Traceback (most recent call last): File...
-
Actually, "confusion" is from my side, and I regret much for that. I believe I have now set all paths to the required locations, executables included, in autogrow_3_1_1.py. What I still dont know how to manage is in #import modules import scipy.optimize The little I know about python is that a python package is a directory of python modules. So, could you help me in treating the above "import scipy.optimize" on the basis that in my OS /usr/lib/python2.7/dist-packages/scipy/optimize/optimize.py and...
-
Hi Francesco. I think I might understand the problem. Please correct me if I'm wrong... Autogrow and Autodock (i.e., MGLTools) are entirely different programs that are similarly named. I am not involved in the Autodock project. For reasons I don't understand, their scripts like prepare_ligand4.py sometimes require the user to use the version of Python they distribute with MGLTools. I imagine they have not included SciPy in that distribution. My program (Autogrow) does require SciPy. So it should...
-
I have now installed Debian 9 (stretch) on a vaio laptop, getting version 18 of python-scipy. Error in running the autogrow tutorial is the same as above descrided. Reexamining what I reported above, it is clear that autogrow is looking for python-scipy from within my MGLTools-latest installation, where there is no python-scipy. From my previous post it is also clear that with Debian python-scipy should be looked for system-wide. In "stretch" /usr/lib/python2.7/dist-packages/scipy/optimize I prefer...
-
It is true that you're using an older version of SciPy, though 0.14.0 does seem to have scipy.optimize: https://docs.scipy.org/doc/scipy-0.14.0/reference/optimize.html I'm using scipy 0.18.1 and had no trouble importing the package: Python 2.7.13 |Anaconda custom (x86_64)| (default, Dec 20 2016, 23:05:08) [GCC 4.2.1 Compatible Apple LLVM 6.0 (clang-600.0.57)] on darwin Type "help", "copyright", "credits" or "license" for more information. Anaconda is brought to you by Continuum Analytics. Please...
-
Debian Python Modules Team python-modules-team@lists.alioth.debian.org Architecture: amd64 Version: 0.14.0-2 Provides: python2.7-scipy as already written. Is that information incomplete/ Thanks for your advice.
-
Thanks for your interest in autogrow. This seems to be a problem with your SciPy installation rather than with autogrow itself. Can you tell me which version of SciPy you have installed? Within Python: import scipy print scipy.version If it's not a very recent version of the module, you might need to upgrade to the latest version. I also recommend using anaconda Python, which often provided access to more recent versions of modules: https://www.continuum.io/downloads Let me know if your SciPy version...
-
Planning to use autogrow, I was trying the tutorial by running ./executable python ../autogrow_3_1_1.py \ -filename_of_receptor 1xdn_receptor.pdb \ -center_x 39.5 -center_y 23.7 -center_z 15.0 \ -size_x 30.0 -size_y 30.0 -size_z 20.0 \ -additional_autoclickchem_parameters "-all_reactions +azide_and_alkyne_to_azole" \ -allow_modification_without_frag_addition FALSE \ -directory_of_source_compounds ./starting_compounds/ \ -directory_of_fragments ../fragments/MW_250/ \ -number_of_mutants_first_generation...
-
I recently got an email from an AutoGrow user who couldn't generate any reactions....
-
jerry kil posted a comment on discussion AutoGrow Support
So i figured it out! I was using a stable source build, which was a bit dated relative...
-
pardon the duplicate post
-
Here is what you requested: OpenBabel version: red@redbox:~$ obabel -V Open Babel...
-
Here is what you requested: OpenBabel version: red@redbox:~$ obabel -V Open Babel...
-
Here is what you requested: OpenBabel version: red@redbox:~$ obabel -V Open Babel...
-
Hi Jerry. Happy to hear you've got AutoGrow installed. Would you mind sending me...
-
Thank you for your prompt reply. i am on my mobile atm, please bear with me. Im running...
-
Hi Jerry. I'm happy you're interested in trying AutoGrow. AutoGrow uses Open Babel...
-
logP is not defined at runtime, not sure why. Any advice? Excited to try this out,...
-
Hi Albert. Please forgive my delay in getting back to you. As currently implemented,...
-
Hi Steve. You aren't by chance using an ancient version of scipy, are you? I think...
-
Steve posted a comment on discussion AutoGrow Support
OK. Iedited the paths in the file. Now I get the following error: comp@inga ~/Apps/autogrow_3_1_1/tutorial...
-
Hi Steve. I'm happy you're planning on using AutoGrow in your research. AutoGrow...
-
I have installed Autogrow v-3.1.1 in Linus Mint v-17.2. and am having a runtime problem....
-
Albert posted a comment on discussion AutoGrow Support
Thank you for such detailed comments. If I understand correctly, if we would like...
-
great, thanks alot. The problem solved now.
-
Hi Albert. Thanks for this question. I did some investigating and was able to figure...
-
Hi Albert. AutoGrow uses a program called AutoClickChem to add fragments to your...
-
Hello: I am using autogrow from new molecules design. In my original protein pdb...
-
Hello: I found that Autogrow is using Autodock Vina for the docking process. However,...
-
Hello: I am just wondering besides introducing "Azide" in the "growing site" of a...
-
Hi Albert. I'll look into adding some other open-source scoring functions (e.g.,...
-
Hi Albert. I'll look into adding some other opensource scoring functions (e.g., RF-SCORE)...
-
Hi, It seems to work. BTW, if possible, could you also implement other free opensource...
-
Hi Albert. If I'm not mistaken, it is possible to restart an AutoGrow job. If you...
-
Hello: I am just wondering can we restart a stopped Autogrow job? For instance: if...
-
Hello: I am using Autogrow for new molecules design these days. It is a really helpful...
-
forget about this thread, it finally works now. I redownload the AUTODOCK tools.
-
Hello: I am running Autogrow with command: python /home/albert/install/autogrow/3.1.1/autogrow.py...
-
Hello: I am running Autogrow with command: python /home/albert/install/autogrow/3.1.1/autogrow.py...
-
Hi Albert. Happy my answers helped. There are two publications: Durrant, J. D., et...
-
I see. Many thanks for such clear explanations. I found this is a really helpful...
-
Hi Albert. The long bond you're seeing is actually an azide group. If you show the...
-
Thank you for the comments. Finally it worked now. I finished the tutorial example...
-
Hi Albert. Not sure, but it could be because you didn't put a space between the "openbabel_bin_directory"...
-
Hi Jdurrant: Many thanks for clear explanations and suggestions. Here I changed the...
-
Hi Albert. I'm happy you're using AutoGrow in your research. I think the problem...
-
Hello: I am using Autogrow these days to design some new compounds which I think...
-
Hi Frits. I'm happy you're interested in AutoGrow. Allow me to answer your questions:...
-
-
-
frits daeyaert posted a comment on discussion AutoGrow Support
Hi Jacob, I've installed autogrow to see how it performs in comparison with my own...
-
You can register for this forum using this link: https://sourceforge.net/user/re...
-
Hi Adam. Happy to hear that the Phenix output PDB does have the right columns. All...
-
Corrected wrong version number in autogrow_3_1_...
-
Minor updates. Now version 3.1.1.
-
Adam Salazar posted a comment on discussion AutoGrow Support
Hi jd, No problem! Since the files didn't look nicely aligned in notepad, I typed...
-
Hi Adam. Thanks for posting your question, and for giving me the idea of starting...
-
Hi jd/AutoGrow support, I'm trying to figure out how to properly format a ligand...
-
-
-
Further minor updates.
-
Minor updates/improvements, version number now ...
1