Please excuse= crossposting.

<= /div>

For example, if one = is doing a QSPR (quantitative structure-properties relation) study of densi= ties of alkanes, and encounters the problem that some densities are measure= d at 20=C2=B0C and others at 25=C2=B0C, how should one handle the inconsist= ency of measurement conditions? Note that the difference in density f= or the same alkane between 20=C2=B0C and 25=C2=B0C might be significant in = comparison to the difference in density between two isomeric alkanes at the= same temperature. Is is legitimate to try to correct/standardize the= 20=C2=B0C densities to 25=C2=B0C densities by subtracting or dividing the = 20=C2=B0C densities by some constant? And if so, how does one determi= ne that constant? Are there other approaches one can use?

--David Shobe

------=_Part_7189248_1879939615.1492812956529-- From owner-chemistry@ccl.net Fri Apr 21 19:12:01 2017 From: "Cristian Bologa CBologa:salud.unm.edu" To: CCL Subject: CCL: What do cheminformaticists do with inconsistently measured data? Message-Id: <-52762-170421191018-6167-Z+govhv+cFrGMa6Et5GKfg^server.ccl.net> X-Original-From: Cristian Bologa Content-Language: en-US Content-Type: multipart/alternative; boundary="_000_149281620388814600saludunmedu_" Date: Fri, 21 Apr 2017 23:10:04 +0000 MIME-Version: 1.0 Sent to CCL by: Cristian Bologa [CBologa(!)salud.unm.edu] --_000_149281620388814600saludunmedu_ Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable David, You might want to have a look at this paper, although I am not sure if that= is what you need. http://pubs.acs.org/doi/abs/10.1021/ie9606451 However, if you have enough overlap between the two datasets (aka same alka= nes measured at both 20 and 25=B0C), I would probably try to derive the rel= ationship between those values myself, using linear or non-linear regressio= n. Cristian ?Cristian Bologa, Ph.D. Research Professor, Div. of Translational Informatics, Dept. of Internal Medicine, Univ. of New Mexico, School of Medicine, Innovation Discovery&Training Center, MSC09 5025, 700 Camino de Salud NE, Albuquerque, NM 87131 tel: +1 (505) 925-7534 fax:+1 (505) 925-7625 ---------------------- "If you never fail, it means you are not trying hard enough" ________________________________ > From: owner-chemistry+cbologa=3D=3Dsalud.unm.edu^-^ccl.net on behalf of David Shobe avidshobe:yahoo= .com Sent: Friday, April 21, 2017 4:15 PM To: Cristian Bologa Subject: CCL: What do cheminformaticists do with inconsistently measured da= ta? Please excuse crossposting. For example, if one is doing a QSPR (quantitative structure-properties rela= tion) study of densities of alkanes, and encounters the problem that some d= ensities are measured at 20=B0C and others at 25=B0C, how should one handle= the inconsistency of measurement conditions? Note that the difference in = density for the same alkane between 20=B0C and 25=B0C might be significant = in comparison to the difference in density between two isomeric alkanes at = the same temperature. Is is legitimate to try to correct/standardize the 2= 0=B0C densities to 25=B0C densities by subtracting or dividing the 20=B0C d= ensities by some constant? And if so, how does one determine that constant= ? Are there other approaches one can use? --David Shobe --_000_149281620388814600saludunmedu_ Content-Type: text/html; charset="iso-8859-1" Content-Transfer-Encoding: quoted-printable

David,

You might want to have a look at this paper, although I am not sure if t= hat is what you need.

http://pubs.ac= s.org/doi/abs/10.1021/ie9606451

However, if you have enough overlap between the two datasets (aka same a= lkanes measured at both 20 and 25=B0C), I would probably try to derive the = relationship between those values myself, using linear or non-linear r= egression.

Cristian

Cristian B= ologa, Ph.D.

Research Professor,
Div. of Translational Informatics,
Dept. of Internal Medicine,
Univ. of New Mexico, School of Medicine,
Innovation Discovery&Training Center, MSC09 5025,
700 Camino de Salud NE, Albuquerque, NM 87131
tel: +1 (505) 925-7534
fax:+1 (505) 925-7625

----------------------

"If you ne= ver fail, it means you are not trying hard enough"

From: owner-chemistry+c= bologa=3D=3Dsalud.unm.edu^-^ccl.net <owner-chemistry+cbologa=3D=3Dsalu= d.unm.edu^-^ccl.net> on behalf of David Shobe avidshobe:yahoo.com <owner-chemistry^-^ccl.net>
Sent: Friday, April 21, 2017 4:15 PM
To: Cristian Bologa
Subject: CCL: What do cheminformaticists do with inconsistently meas= ured data?

Please excuse crosspostin= g.

For example, if one is do= ing a QSPR (quantitative structure-properties relation) study of densities = of alkanes, and encounters the problem that some densities are measured at = 20=B0C and others at 25=B0C, how should one handle the inconsistency of measurement conditions? Note that th= e difference in density for the same alkane between 20=B0C and 25=B0C might= be significant in comparison to the difference in density between two isom= eric alkanes at the same temperature. Is is legitimate to try to correct/standardize the 20=B0C densities to 25=B0C= densities by subtracting or dividing the 20=B0C densities by some constant= ? And if so, how does one determine that constant? Are there ot= her approaches one can use?

--David Shobe

--_000_149281620388814600saludunmedu_--