Dear participants! \nI want to say one strange thing data in xyz structure. If you convert this data from xyz format to mol via following instruction: http://proteinsandwavefunctions.blogspot.com/2018/01/xyz2mol-converting-xyz-file-to-rdkit.html, you receive QM9 dataset.
", "messages": [ { "id": 550955, "postDate": "2019-06-12T07:12:34.757Z", "content": "There is no leak. Please update your title as it is merely click-bait written in caps.
", "rawMarkdown": "There is no leak. Please update your title as it is merely click-bait written in caps.", "votes": 3 }, { "id": 550584, "postDate": "2019-06-11T19:13:01.620Z", "content": "well.. tht is not true.. as the name xyz2mol suggests, it only converts the structures and does not provide quantum data. (edit: it also provides bond orders..)
", "rawMarkdown": "well.. tht is not true.. as the name xyz2mol suggests, it only converts the structures and does not provide quantum data. (edit: it also provides bond orders..)", "votes": 2 }, { "id": 550565, "postDate": "2019-06-11T18:47:48.470Z", "content": "Dear participants! \nI want to say one strange thing data in xyz structure. If you convert this data from xyz format to mol via following instruction: http://proteinsandwavefunctions.blogspot.com/2018/01/xyz2mol-converting-xyz-file-to-rdkit.html, you receive QM9 dataset.
", "rawMarkdown": "Dear participants! \nI want to say one strange thing data in xyz structure. If you convert this data from xyz format to mol via following instruction: http://proteinsandwavefunctions.blogspot.com/2018/01/xyz2mol-converting-xyz-file-to-rdkit.html, you receive QM9 dataset. \n", "votes": -3 }, { "id": 550614, "postDate": "2019-06-11T20:22:45.540Z", "content": "I run the following to make it work
\n\n```\nconda install -c conda-forge rdkit\npip install networkx
\n\n```\nand in jupyter
\n\n```\n%run xyz2mol --sdf test_000001.xyz
\n\n```\nwhich returns SMILES code and creates a file test_000001.sdf. However it contains some weird fields, specifically
\n\n```\n RDKit 3D
\n\n5 4 0 0 0 0 0 0 0 0999 V2000\n -0.0127 1.0858 0.0080 C 0 0 0 0 0 0 0 0 0 0 0 0\n 0.0022 -0.0060 0.0020 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1.0117 1.4638 0.0003 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5408 1.4475 -0.8766 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5238 1.4379 0.9064 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1 2 1 0\n 1 3 1 0\n 1 4 1 0\n 1 5 1 0\nM END\n$$$$
\n\n```
\n\nany suggestions?
\n\nUPDATE: I removed the comparison with QM9, apparently even that is not allowed.
", "rawMarkdown": "I run the following to make it work\n\n```\nconda install -c conda-forge rdkit\npip install networkx\n\n```\nand in jupyter\n\n```\n%run xyz2mol --sdf test_000001.xyz\n\n```\nwhich returns SMILES code and creates a file test_000001.sdf. However it contains some weird fields, specifically\n\n```\n RDKit 3D\n\n 5 4 0 0 0 0 0 0 0 0999 V2000\n -0.0127 1.0858 0.0080 C 0 0 0 0 0 0 0 0 0 0 0 0\n 0.0022 -0.0060 0.0020 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1.0117 1.4638 0.0003 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5408 1.4475 -0.8766 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5238 1.4379 0.9064 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1 2 1 0\n 1 3 1 0\n 1 4 1 0\n 1 5 1 0\nM END\n$$$$\n\n```\n\nany suggestions?\n\nUPDATE: I removed the comparison with QM9, apparently even that is not allowed. ", "votes": -1, "replies": [ { "id": 550617, "postDate": "2019-06-11T20:38:42.747Z", "content": "Even after being told off for sharing QM9, knowing breaking the rules of the competition, and being told by the Kaggle team that your approach was extremely poor form, you continue to use and share your methods with QM9? @inversion @larsbratholm
", "rawMarkdown": "Even after being told off for sharing QM9, knowing breaking the rules of the competition, and being told by the Kaggle team that your approach was extremely poor form, you continue to use and share your methods with QM9? @inversion @larsbratholm ", "votes": -2 }, { "id": 550620, "postDate": "2019-06-11T20:46:10.667Z", "content": "What? I am clearly not using it here for validating or developing my models, what are you talking about. Read the rules again.
", "rawMarkdown": "What? I am clearly not using it here for validating or developing my models, what are you talking about. Read the rules again." }, { "id": 550621, "postDate": "2019-06-11T20:47:43.627Z", "content": "Why else would you be using it...
", "rawMarkdown": "Why else would you be using it...", "votes": -3 }, { "id": 550633, "postDate": "2019-06-11T21:04:33.930Z", "content": "Ok, @matthewmasters , you are probably right, given that inversion liked your post. Posting a single entry from QM9 for comparison is not allowed here as well. Removed it.
", "rawMarkdown": "Ok, @matthewmasters , you are probably right, given that inversion liked your post. Posting a single entry from QM9 for comparison is not allowed here as well. Removed it." }, { "id": 550949, "postDate": "2019-06-12T06:57:12.643Z", "content": "The extra fields are just guesses whether the bonds in the molecules are double, single or triple bonds. Someone with a basic level of chemistry understanding and the SDF-file specifications at hand will be able to recreate those numbers by hand, and they do not involve any QM calculations.
", "rawMarkdown": "The extra fields are just guesses whether the bonds in the molecules are double, single or triple bonds. Someone with a basic level of chemistry understanding and the SDF-file specifications at hand will be able to recreate those numbers by hand, and they do not involve any QM calculations.", "votes": 1 } ] }, { "id": 550573, "postDate": "2019-06-11T18:58:00.753Z", "content": "OMG !!\nNow we know why it was banned to use.
", "rawMarkdown": "OMG !!\nNow we know why it was banned to use.", "replies": [ { "id": 550574, "postDate": "2019-06-11T19:02:11Z", "content": "I'm not sure whether it makes sense to ban some external data, if we can reconstruct it from the available data (whish is definitely within rules) ...
", "rawMarkdown": "I'm not sure whether it makes sense to ban some external data, if we can reconstruct it from the available data (whish is definitely within rules) ...", "votes": 2 }, { "id": 550575, "postDate": "2019-06-11T19:04:28.730Z", "content": "I did only 2 submission. I was very exited about this comptetion. Lol
", "rawMarkdown": "I did only 2 submission. I was very exited about this comptetion. Lol" }, { "id": 550597, "postDate": "2019-06-11T19:42:14.470Z", "content": "@artgor Reconstruct what exactly? The quantum properties provided in the qm dataset? I don't think so...
", "rawMarkdown": "@artgor Reconstruct what exactly? The quantum properties provided in the qm dataset? I don't think so..." } ] }, { "id": 550783, "postDate": "2019-06-12T03:05:33.843Z", "content": "It is obvious that xyz files contain only information about kinds of atoms and their coordinates, we won't get quantum mechanical information from them, if we deal with them without any special tools.\nWhat's hidden is not a leak in data from xyz files, but the existence of a special tool supplying quantum mechanical information.
", "rawMarkdown": "It is obvious that xyz files contain only information about kinds of atoms and their coordinates, we won't get quantum mechanical information from them, if we deal with them without any special tools.\nWhat's hidden is not a leak in data from xyz files, but the existence of a special tool supplying quantum mechanical information.", "votes": 1, "isDeleted": true }, { "id": 550596, "postDate": "2019-06-11T19:41:05.357Z", "rawMarkdown": "", "votes": -1, "isDeleted": true } ], "comments": [ { "id": 550955, "author_name": "ANDERSX", "author_url": "", "post_date": "2019-06-12T07:12:34.757000", "content": "There is no leak. Please update your title as it is merely click-bait written in caps.
", "votes": 3, "replies": [] }, { "id": 550584, "author_name": "skhan", "author_url": "", "post_date": "2019-06-11T19:13:01.620000", "content": "well.. tht is not true.. as the name xyz2mol suggests, it only converts the structures and does not provide quantum data. (edit: it also provides bond orders..)
", "votes": 2, "replies": [] }, { "id": 550614, "author_name": "nosound", "author_url": "", "post_date": "2019-06-11T20:22:45.540000", "content": "I run the following to make it work
\n\n```\nconda install -c conda-forge rdkit\npip install networkx
\n\n```\nand in jupyter
\n\n```\n%run xyz2mol --sdf test_000001.xyz
\n\n```\nwhich returns SMILES code and creates a file test_000001.sdf. However it contains some weird fields, specifically
\n\n```\n RDKit 3D
\n\n5 4 0 0 0 0 0 0 0 0999 V2000\n -0.0127 1.0858 0.0080 C 0 0 0 0 0 0 0 0 0 0 0 0\n 0.0022 -0.0060 0.0020 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1.0117 1.4638 0.0003 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5408 1.4475 -0.8766 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5238 1.4379 0.9064 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1 2 1 0\n 1 3 1 0\n 1 4 1 0\n 1 5 1 0\nM END\n$$$$
\n\n```
\n\nany suggestions?
\n\nUPDATE: I removed the comparison with QM9, apparently even that is not allowed.
", "votes": -1, "replies": [ { "id": 550617, "author_name": "Matt", "author_url": "", "post_date": "2019-06-11T20:38:42.747000", "content": "Even after being told off for sharing QM9, knowing breaking the rules of the competition, and being told by the Kaggle team that your approach was extremely poor form, you continue to use and share your methods with QM9? @inversion @larsbratholm
", "votes": -2, "replies": [] }, { "id": 550620, "author_name": "nosound", "author_url": "", "post_date": "2019-06-11T20:46:10.667000", "content": "What? I am clearly not using it here for validating or developing my models, what are you talking about. Read the rules again.
", "votes": 0, "replies": [] }, { "id": 550621, "author_name": "Matt", "author_url": "", "post_date": "2019-06-11T20:47:43.627000", "content": "Why else would you be using it...
", "votes": -3, "replies": [] }, { "id": 550633, "author_name": "nosound", "author_url": "", "post_date": "2019-06-11T21:04:33.930000", "content": "Ok, @matthewmasters , you are probably right, given that inversion liked your post. Posting a single entry from QM9 for comparison is not allowed here as well. Removed it.
", "votes": 0, "replies": [] }, { "id": 550949, "author_name": "ANDERSX", "author_url": "", "post_date": "2019-06-12T06:57:12.643000", "content": "The extra fields are just guesses whether the bonds in the molecules are double, single or triple bonds. Someone with a basic level of chemistry understanding and the SDF-file specifications at hand will be able to recreate those numbers by hand, and they do not involve any QM calculations.
", "votes": 1, "replies": [] } ] }, { "id": 550573, "author_name": "Manoj", "author_url": "", "post_date": "2019-06-11T18:58:00.753000", "content": "OMG !!\nNow we know why it was banned to use.
", "votes": 0, "replies": [ { "id": 550574, "author_name": "Andrey Lukyanenko", "author_url": "", "post_date": "2019-06-11T19:02:11", "content": "I'm not sure whether it makes sense to ban some external data, if we can reconstruct it from the available data (whish is definitely within rules) ...
", "votes": 2, "replies": [] }, { "id": 550575, "author_name": "Manoj", "author_url": "", "post_date": "2019-06-11T19:04:28.730000", "content": "I did only 2 submission. I was very exited about this comptetion. Lol
", "votes": 0, "replies": [] }, { "id": 550597, "author_name": "Matt", "author_url": "", "post_date": "2019-06-11T19:42:14.470000", "content": "@artgor Reconstruct what exactly? The quantum properties provided in the qm dataset? I don't think so...
", "votes": 0, "replies": [] } ] }, { "id": 550783, "author_name": "", "author_url": "", "post_date": "2019-06-12T03:05:33.843000", "content": "It is obvious that xyz files contain only information about kinds of atoms and their coordinates, we won't get quantum mechanical information from them, if we deal with them without any special tools.\nWhat's hidden is not a leak in data from xyz files, but the existence of a special tool supplying quantum mechanical information.
", "votes": 1, "replies": [] }, { "id": 550596, "author_name": "", "author_url": "", "post_date": "2019-06-11T19:41:05.357000", "content": "", "votes": -1, "replies": [] } ], "raw_markdown_by_id": { "550955": "There is no leak. Please update your title as it is merely click-bait written in caps.", "550584": "well.. tht is not true.. as the name xyz2mol suggests, it only converts the structures and does not provide quantum data. (edit: it also provides bond orders..)", "550565": "Dear participants! \nI want to say one strange thing data in xyz structure. If you convert this data from xyz format to mol via following instruction: http://proteinsandwavefunctions.blogspot.com/2018/01/xyz2mol-converting-xyz-file-to-rdkit.html, you receive QM9 dataset. \n", "550614": "I run the following to make it work\n\n```\nconda install -c conda-forge rdkit\npip install networkx\n\n```\nand in jupyter\n\n```\n%run xyz2mol --sdf test_000001.xyz\n\n```\nwhich returns SMILES code and creates a file test_000001.sdf. However it contains some weird fields, specifically\n\n```\n RDKit 3D\n\n 5 4 0 0 0 0 0 0 0 0999 V2000\n -0.0127 1.0858 0.0080 C 0 0 0 0 0 0 0 0 0 0 0 0\n 0.0022 -0.0060 0.0020 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1.0117 1.4638 0.0003 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5408 1.4475 -0.8766 H 0 0 0 0 0 0 0 0 0 0 0 0\n -0.5238 1.4379 0.9064 H 0 0 0 0 0 0 0 0 0 0 0 0\n 1 2 1 0\n 1 3 1 0\n 1 4 1 0\n 1 5 1 0\nM END\n$$$$\n\n```\n\nany suggestions?\n\nUPDATE: I removed the comparison with QM9, apparently even that is not allowed. ", "550573": "OMG !!\nNow we know why it was banned to use.", "550783": "It is obvious that xyz files contain only information about kinds of atoms and their coordinates, we won't get quantum mechanical information from them, if we deal with them without any special tools.\nWhat's hidden is not a leak in data from xyz files, but the existence of a special tool supplying quantum mechanical information.", "550596": "" } }