tag:blogger.com,1999:blog-7844526396210378482.comments2024-01-31T09:23:26.925+00:00Noel O'BlogNoel O'Boylehttp://www.blogger.com/profile/03288289351940689018noreply@blogger.comBlogger1152125tag:blogger.com,1999:blog-7844526396210378482.post-3600023577105659862023-10-20T03:23:59.977+01:002023-10-20T03:23:59.977+01:00Hi Noel, thanks for sharing the code. I have probl...Hi Noel, thanks for sharing the code. I have problem running the codes.<br /><br />amax = np.amax(adist)<br />TypeError: '>=' not supported between instances of 'map' and 'map'<br /><br /><br />results = mds.fit(adist)<br />TypeError: float() argument must be a string or a real number, not 'map'<br /><br />I am using python 3.7Maggie Don Lauhttps://www.blogger.com/profile/08566880432332268302noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-7735803159508933932023-09-23T16:45:37.896+01:002023-09-23T16:45:37.896+01:00Neat topic, it was great to talk in Mainz!Neat topic, it was great to talk in Mainz!Charles Tapley Hoythttps://www.blogger.com/profile/07076861594720975062noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-48942306125706367962023-06-27T20:50:26.584+01:002023-06-27T20:50:26.584+01:00Max is indeed present in 2 enantiomers.Max is indeed present in 2 enantiomers.Peter Ertlhttps://www.blogger.com/profile/02620509126764627686noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-53991617195819285932022-08-04T15:48:12.375+01:002022-08-04T15:48:12.375+01:00Stay tuned for the next blogpost.Stay tuned for the next blogpost.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-63531854682758704892022-08-03T09:34:25.753+01:002022-08-03T09:34:25.753+01:00Would DeepSMILES be a better solution to this prob...Would DeepSMILES be a better solution to this problem than SELFIES? ThanksAnonymousnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-86662705294552431982022-06-03T16:18:22.093+01:002022-06-03T16:18:22.093+01:00Note also that computational chemistry is a broade...Note also that computational chemistry is a broader topic than cheminformatics and is closer to the drug discovery process in some ways, in that it involves directly using protein structural information to help guide a medicinal chemistry team. This includes homology modelling, protein-ligand docking and molecular dynamics among other techniques. If this is of interest, you should reach out to an existing group that is publishing in these areas and chat about this.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-52517227425973499222022-05-30T11:51:29.281+01:002022-05-30T11:51:29.281+01:00RSC CICAG have an expanding selection of YouTube v...RSC CICAG have an expanding selection of YouTube videos describing Open-Source tools and resources that would be worth watching https://www.youtube.com/c/RSCCICAG<br />Chrishttps://www.blogger.com/profile/11553312663186781938noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-52859240877593792472022-05-30T08:45:36.032+01:002022-05-30T08:45:36.032+01:00I studied Biology for my undergraduate and it was ...I studied Biology for my undergraduate and it was not until my MSc in Bonn (Life Science Informatics) that I learnt about chemoinformatics. There are very few places that have formal education on chemoinformatics, the University of Bonn is one of the few I know. Another great resource are the summer schools in Strassbourg organized by Prof. Varnek. There used to be a MSc in the UK but it no longer runs. <br /><br />But as many people have said, it is not needed and many start when they do a PhD that is related to the topic. Knowing Python already is a great start, if you haven't already learn how to use RDKit, it is a great chemoinformatics toolkit. I also second @Nils comment on machine learning, it is a great technique to know about in the current job market. If you are interested in a research career, doing a PhD is almost necessary whether you want to go to academia or industry. <br /><br />Probably someone who would be great for you to get in touch with would be Prof. Jose Medina-Franco, he does chemoinformatics at the University of Mexico. He would probably know more about the chemoinformatics scene in Latin America. He is in twitter under https://twitter.com/difacquim.Antonio de la Vega de Leonhttp://adlvdl.githb.ionoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-72229994205423436202022-05-29T18:16:51.432+01:002022-05-29T18:16:51.432+01:00I think Drew Conway's Data Science Venn Diagra...I think Drew Conway's <a href="http://drewconway.com/zia/2013/3/26/the-data-science-venn-diagram" rel="nofollow">Data Science Venn Diagram</a> provides solid guidance for a number of fields, including Cheminformatics. According to Conway, one needs three skills. <br /><br />- Domain knowledge - In Cheminformatics, we work with chemical and biological data. In order to determine whether our models are meaningful, we need to understand the underlying science. For those coming from computer science, it's useful to gain some background in organic chemistry and molecular biology. <br /><br />- Hacking skills - In order to do Cheminformatics, you need to know how to code. One way to get better at coding is to study and modify code from blog posts or papers. The Journal of Cheminformatics and The Journal of Chemical Information and Modeling have lots of papers with code. <br /><br />- A knowledge of math and statistics. A lot of modern Cheminformatics involves building machine learning models. A background in statistics enables you to be able to appropriately validate and compare models. <br /><br />I've put together a set of <a href="https://github.com/PatWalters/practical_cheminformatics_tutorials" rel="nofollow">Jupyter notebooks</a> that I think could provide a good starting point for someone learning Cheminformatics. Finally, ask a lot of questions, there are many enthusiastic Cheminformaticians out there who would be happy to help. <br />Pat Waltershttps://www.blogger.com/profile/18223198920629617711noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-75636723717591995432022-05-29T18:07:44.617+01:002022-05-29T18:07:44.617+01:00First of all: congratulations on this great career...First of all: congratulations on this great career choice! There are plenty of interesting topics in this field, but the "hottest" ones seem to be around applying machine learning / deep learning to problems in chemistry. You should find plenty of articles on predicting molecular properties or generating structures with a given target profile in journals. If I had the chance to start a new PhD, I would probably look at predicting chemical reactions (regioselectivity, optimal conditions, yields) and how to apply this to retrosynthesis. There is a shortage of data on this topic in the public domain, but you might be able to generate your own training data by using classical computational chemistry approaches (e.g. DFT) to calculate what you need. Nils Weskampnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-81998462941922841922022-05-24T23:27:44.877+01:002022-05-24T23:27:44.877+01:00Hi! I am a third-year undergrad student studying c...Hi! I am a third-year undergrad student studying chemistry with a biology minor. I have been thinking to become a pharmacologist, but the time goes by I discovered that I am more interested in computational chemistry. And I have been doing undergrad research since freshman year in computational chemistry which kindly made me fall in love with computational chemistry than a wet lab. And I figured cheminformatics can help me to work on what I love designing/discovering drugs without becoming a pharmacologist and still apply computer and chemistry knowledge. I am trying to figure out what to study in Graduate school now that could lead me to become a cheminformatics. Do you have any suggestions on what would be best to do a Ph.D. on? Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-23176274365409838352021-12-03T17:46:14.514+00:002021-12-03T17:46:14.514+00:00It's linked in Felix's blog post: https://...It's linked in Felix's blog post: https://homepage.univie.ac.at/felix.plasser/chemprogs/python.htmNoel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-83123943912698661672021-12-03T15:22:01.388+00:002021-12-03T15:22:01.388+00:00where is this code?where is this code?Anonymousnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-35962683357773932122021-04-29T11:33:00.795+01:002021-04-29T11:33:00.795+01:00It's part of Open Babel now.It's part of Open Babel now.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-45125645534643563442021-04-28T16:25:09.279+01:002021-04-28T16:25:09.279+01:00Dear Prof,
I can not download the code, because t...Dear Prof, <br />I can not download the code, because the link is not available.<br />Could you please provide me your code via email (daoduyquang-at-gmail.com)<br />Thanks so much, <br />QuangAnonymoushttps://www.blogger.com/profile/11625478593276405080noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-68843682063248304842021-02-17T21:15:07.640+00:002021-02-17T21:15:07.640+00:00There is one edge case which is not considered: wh...There is one edge case which is not considered: when Boron atoms have 4 covalent bonds , it has a negative charge and no positive charge around it so the alg tries tries to add 0 - (-1) =1H however this will mean Boron now has 5 covalent bonds, which is not chemically possible. Check for this molecule : N[C@@H](CCCC[B-](O)(O)O)C(=O)OAnonymoushttps://www.blogger.com/profile/07125547586991260918noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-20305772302415183072020-10-16T15:49:26.467+01:002020-10-16T15:49:26.467+01:00Very nice, N. I need to think this over but clearl...Very nice, N. I need to think this over but clearly it's touching on the same issues.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-52078046520442457432020-10-16T15:35:26.797+01:002020-10-16T15:35:26.797+01:00https://en.wikipedia.org/wiki/Arrow%27s_impossibil...https://en.wikipedia.org/wiki/Arrow%27s_impossibility_theoremNhttps://www.blogger.com/profile/13458388727557070416noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-1862383570858682242020-10-09T20:52:06.799+01:002020-10-09T20:52:06.799+01:00No version of the average will help as they allow ...No version of the average will help as they allow methods C, D, etc. to influence the question of whether A > B.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-80045481956854444122020-10-09T20:19:10.326+01:002020-10-09T20:19:10.326+01:00Would using the median of the rank solve some of t...Would using the median of the rank solve some of these problems?Jamesnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-55757841178289048762020-07-08T03:15:38.187+01:002020-07-08T03:15:38.187+01:00I think cheminformatics is a new platform with bri...I think cheminformatics is a new platform with brighter future. One of the advantages is that no worry on the ethical issues when doing experiments as well as getting involved in false positive or false negative results. Once proven computational results, we may proceed to prove in wet lab. This would be safer in terms of risks and costAnonymoushttps://www.blogger.com/profile/15025591282609326376noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-23898956191438222752020-06-08T16:54:54.041+01:002020-06-08T16:54:54.041+01:00This post is 9 years old. I don't remember wha...This post is 9 years old. I don't remember what the idea is behind step 3. It must have been to get newer versions of some of the source files, and so this may no longer be necessary. Those links were to the source code repository of PyRx, and are now broken.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-26515666175969218402020-06-08T14:39:58.367+01:002020-06-08T14:39:58.367+01:00Dear Author,
Thank you for this tutorial.
What do ...Dear Author,<br />Thank you for this tutorial.<br />What do you mean by item (3): Replace the appropriate file in "C:\Program Files (x86)\PyRx\Lib\site-packages\PyRx" with importWizard.py, dbPlot.py and autodockNavigator.py.<br /><br />After install the files exists?<br />Thanks,<br />KarimKarimnoreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-35257449321572259162020-04-09T16:54:43.897+01:002020-04-09T16:54:43.897+01:00It's an oldie but a goodie.It's an oldie but a goodie.Noel O'Boylehttps://www.blogger.com/profile/03288289351940689018noreply@blogger.comtag:blogger.com,1999:blog-7844526396210378482.post-27065285189031871902020-04-09T16:36:25.003+01:002020-04-09T16:36:25.003+01:00We just hit 3,000 citations on Google Scholar and ...We just hit 3,000 citations on Google Scholar and over 2,000 by whatever they use on the site, so another outlier from Avogadro seems like a solid guess. Just saw this post on the altmetrics analysis.Marcus D. Hanwellhttps://www.blogger.com/profile/08187423171219662136noreply@blogger.com