Errata

Scientific publications Philippe Hünenberger: typos and minor errors

This page provides a list of typos and minor errors discovered in my publications after printing (please, don't hesitate to contact me if you find others); two articles, [17] and [55], were affected by more significant errors, justifying the publication of an erratum.

The numbering refers to my publication list

[10] Hünenberger & van Gunsteren, J. Chem. Phys. 108 6117 (1998)

• Page 6126, Figure 2a: The labels of the curves CG/1 and AT/2 are interverted (the solid line is CG/1 and the dashed line is AT/2), see [132] (page 26423) where the error was reported.
  

[14] Hünenberger & McCammon, J. Chem. Phys. 110 1856 (1999)

• Page 1868, Eq(A11): δ_j,I should be replaced by δ_j,J and δ_k,I by δ_k,K
• Page 1868, Eq(A11) is incorrect as written: the 6 terms in lines 2-4 of Eq(A6) should be added to the left-hand-side of this equation

[17] Baker et al., J. Chem. Phys. 110 10679 (1999)

• Page 10688, Eq(66): the sign of the last term in the right-hand side should be minus instead of plus, see our book p 79 [Eq(3.42) and following lines] where this error was reported
• See also Erratum: J. Chem. Phys. 113 2510 (2000)

[18] Hünenberger, In "Simulation and theory of electrostatic interactions in solution..." pp 17-83 (1999)

• Page 27, Figure 4: The labels of the LΨ and LΨ_η curves should be swapped, i.e. LΨ_η is the solid line and LΨ the short-dashed line
• Page 34, 9 lines below header "Shaping functions": -2.837927 should be replaced by -2.837297

[22] Heinz et al., J. Chem. Phys. 115 1125 (2001)

• Page 1127, 4 lines below Eq(14): p~_α(nΔN) should read p~_α(nΔM)

[25] Börjesson & Hünenberger, J. Chem. Phys. 114 9706 (2001)

• Page 9708, Eq(6): in the right-hand side of the equation, the sum over the 2^N vectors l (all possible N-dimensional vectors with components equal to 0 or 1) should actually involve terms that are products from i=1 to N of the indicated quantity
• Page 9708, Eq(6): this equation is also somewhat clumsy, because it implies the convention 0⁰=1 (this case can occur when λ_i=0 or λ_i=1)
• Page 9708, Eq(6): finally, the Hamiltonian H in the right-hand side should probably rather be called H_phy, and corresponds to physical (0 or 1) protonation states (otherwise, it seems that H_tot is used in this equation to define itself)
• This article has been criticised by Baptista, J. Chem. Phys. 116 7766 (2002); we think at present that the criticism is justified, and agree that the proposed constant-pH method is empirical in nature (mean-field approximation), rather than formally rigorous, as initially thought

[37] Pereira et al., Biophys. J. 86 2273 (2004)

• Table 1, 3 entries at 475K: the reported values for D_xy and D_z should be divided by 10, see [56] p 15575 (Table 1, footnote a) where this error was reported
• Figure 4, graph legend: CH₂ Glycerol should read CH₁ Glycerol

[41] Hünenberger, Adv. Polymer. Sci. 173 105 (2005)

• Table 1, entry μPR: the name of the ensemble should be "grand-isoenthalpic-isobaric" and not "grand-isothermal-isobaric"
• Page 114, Eq(12): the quantity summed over i in the right-hand side should read m_i[(r_i-r_CM)²1 - (r_i-r_CM) ⊗(r_i-r_CM)] instead of m_i(r_i-r_CM) ⊗(r_i-r_CM)
• Page 128, Eq(56): the prefactor in the right-hand-side should be -1/2 instead of 1/2
• Page 128, Eq(57): the prefactor of the second term in the right-hand-side should be +1/2 instead of -1/2
• Page 143, Eq(113), line 3: left-hand-side, in -∂ρ/∂γ⋅γ, the second γ should be γ-dot (dot overscript)

[47] Heinz & Hünenberger, J. Chem. Phys. 123 034107 (2005)

• Page 4, Eq(9): the second sum should involve index j instead of i
• Page 5, 6 lines below Eq(25): -2.837927 should be replaced by -2.837297
• Page 7, Eq(41): the second sum should involve index j instead of i
• Page 9, Eq(65): the quantity k³R_RF in the denominator of one of the fractions (second line) should actually be k³R³_RF
• Page 9, Eq(65): the last condition k≠0 (bottom of the second line) should actually be k=0
• Page 18, Eq(C1): e^-κ2/4 should be replaced by e^-κ2/4

[49] Christen et al., J. Comput. Chem. 26 1719 (2005)

• Table 7, entry p=3: w_p(ξ) should read 3/4-ξ² instead of -(3/4)(4ξ²-1)

[50] Kastenholz & Hünenberger, J. Chem. Phys. 124 124106 (2006)

• Page 8, Point (1): the estimated value of ΔΦ_ODL for liquid argon at 87K based on the Slater-type analysis should be -411 kJ mol^-1e^-1 (-4.26 V) instead of -463 kJ mol^-1e^-1 (-4.80 V), see below
• Supplementary Material, Eq(S2.2): η should be 21.6 instead of 24.3 nm^-3, and ΔΦ^QM_ODLshould be -411 instead of -463 kJ mol^-1e^-1, the reason is that the molar mass of argon was erroneously set to 35.45 instead of 39.948 g mol^-1, see our book p 283 [Eq(4.342) and following lines], 284 [Eq(4.347) and following lines] and 285, where the correct results are reported

[51] Kastenholz & Hünenberger, J. Chem. Phys. 124 224501 (2006)

• Table III, entries LS/2048, LA/2048 and LV/2048: box edge (3^rd column) should be 3.94 instead of 3.13 nm
• Page 7, 4 lines below Eq(39): one could expect the correction term to be proportional to (1-ε_S^-1), and not (1-ε_S)^-1 as written; however, even in this altered form, the statement is not correct, see [85] p 5 [comment just below Eq(15)] or our book p 504 (comment just below Table 6.2), where the statement is ammended

[55] Pereira et al., Biophys. J. 90 4337 (2006)

• Figure 1, drawing for Nigerose (N): the arrow marked 49.5% should point in the opposite direction, see Table 1 where the H-bond considered is (correctly) reported in the direction HO₄'→O₂
• Table 1, caption, lines 1-2: the ω' values should read gg -60° and gt +60°, see [79] pp 1784-1786 and 1790 as well as [84] pp 1002-1004 for precise (and correct) definitions
• See also Erratum: Biophys. J. 93 706 (2007)
• The erratum also contains a serious mistake in its Table 1: the entries ΔS_cl^ah and ΔS_cl^pc should be multiplied by 8.314, and the entries S^ctd and -TΔS^ctd recalculated accordingly, see [79] p 1796 where this error was reported

[57] Baron et al., Trends Phys. Chem. 11 87 (2006)

• Page 118, Eq(C5): the left-hand-side of this equation should read ln κ₁^o instead of κ₁^o
• Page 118, Eq(C6): the left-hand-side of this equation should read ln κ₂^o instead of κ₂^o

[66] Perić et al., 34 421 (2008)

• Abstract, line 3: abreviation for poly-α(1→4)-L-guluronate should read pGulU instead of pGlcU

[65] Kräutler & Hünenberger, Mol. Simul. 34 491 (2008)

• Figure 4: the radial distribution functions are incorrectly normalized, probably to unity (three bottom panels) or half-unity (top panel) instead of the box volume V_box (i.e. the integral of 4πr²g(r) is one or one-half instead of V_box); the reported g(r) have thus units of nm^-3 and must be scaled by V_box (three bottom panels) or 2V_box (top panel) with V_box = 287.5 (CU box) or 228.3 (TO box) nm³ to obtain properly normalized (unitless) radial distribution functions.

[81] Horta et al., J. Chem. Theory Comput. 6 2488 (2010)

• Page 2491, 15 lines below Table 1: a value of 4.575·10^-4 (kJ·mol^-1·nm^-3)^-1 was used for the simulations, but it is not the value appropriate for pure water, which would be 7.505·10^-4 (kJ·mol^-1·nm^-3)^-1 based on a compressibility of 4.52·10^-5 bar^-1 at 25°C; the value used is the approximate value suggested in the GROMOS96 manual for a 1:1 protein-water mixture
• Page 2493, Table 3: the numbers reported in this table and discussed in the text for N_LL^intra and N_WW are actually not the average numbers of intralayer lipid-lipid and water-water H-bonds, respectively, but twice these numbers; given this doubling, the scaling of N_LL^intra by the number of lipids in the system provides the average number of lipids H-bonded to a given lipid within a leaflet, and the discussion on page 2493 (right column, second paragraph) is correct; note that unlike N_LL^intra, N_LL^inter is not doubled, i.e. it is the average number of interlayer lipid-lipid H-bonds; this is a nomenclature issue and not a mistake, see [99] for a more consistent formulation
• Page 2493, Table 3: in addition to the previous nomenclature issue, the numbers reported in this table for N_LL^intra (i.e. twice the average number of intralayer lipid-lipid H-bonds) are affected by a small calculation mistake; the correct values can be obtained by subtracting N_LL^inter from these numbers; the error is essentially negligible (with the possible exception of the three last entries at quarter hydration) and does not affect the main conclusions of the discussion on page 2493 (right column, second paragraph)
• Page 2497, Figure 5, points at n=2: the values for these points are affected by a small calculation error and are too low (this problem also affects [99], Figure 3), see [104] (caption of Figure 9) and [105] (caption of Figure 8) where the error was reported and correct values for these points can be found

[82] Hansen et al., J. Chem. Theory Comput. 6 2596 (2010)

• Page 2617, 6 lines below Eq (A.8): k_LE is equal to (2πσ)^N_LE/2c_LE and not (2πσ)^N_LE/2c_LE

[83] Hansen & Hünenberger, J. Chem. Theory Comput. 6 2622 (2010)

• Page 2627, 4 lines above Eq (5): W_l should be referred to as a "half" width instead of a "double" width, i.e. the width of the "tube" defined by the flat-bottom region of the transverse biasing potential is 2W_l
• Page 2631, Eq(27): this equation as well as the following descriptive lines are formally incorrect (although the calculations were performed correctly); <X(r)>_Q(r)∈S should be replaced by <X(r)θ(r)> where θ(r) is one if Q(r)∈S and zero otherwise, and the averaging is performed over the entire biased ensemble; since C_G can be chosen arbitrarily, this is equivalent to an integral (and not an average) over the region for which Q(r)∈S

[84] Hansen & Hünenberger, J. Comput. Chem. 32 998 (2011)

• Table 2, last line, improper dihedrals for β-Glc: C₂-C₁-O₁-C₂ must be changed to C₂-C₃-C₁-O₂, and C₅-O₅-C₆-C₄ to C₅-C₄-O₅-C₆
• Table 3, entry C6: 45A4/53A6 as well as 56A_CARBO atom types are incorrect and should both be CH₂
• Tables 3 and 4: the atom type HO actually does not exist in GROMOS - in the force-field files, it is listed simply as H
• Table 6, entry T1, dihedrals for β-Glc: O₅-C₅-C₆-O₆ should rather be defined as C₄-C₅-C₆-O₆ to follow the the IUPAC definition of ω; this does not matter much in practice since T1 is a 3-fold term (the definition C₄-C₅-C₆-O₆ is adopted in the current GROMOS force-field files)
• Table 6, entry T3, dihedrals for β-Glc: C₅-C₄-O₄-H₄ should rather be defined as C₃-C₄-O₄-H₄ to follow the systematism of the other dihedrals and the IUPAC definition of χ₄; this does not matter much in practice since T3 is a 3-fold term (the definition C₃-C₄-O₄-H₄ is adopted in the current GROMOS force-field files)
• Table 8, Gal ω-rotamers, entry for 45A4(β) from Ref 73: the populations should be 34, 41 and 25 (instead of 34, 41 and 35)

[85] Reif & Hünenberger, J. Chem. Phys. 134 144103 (2011)

• Page 7, Eq. 32: This equation is formally not correct, considering that q_I=0 in the situation considered here. One should write instead: χ^ion = Φ_o^raw + [2/(N_A q_I) ΔG_B^LS]^* + [1/(N_A q_I) ΔG_C1^LS]^*, where the "*" superscript indicates that the quantity is simplified assuming q_I≠0 and subsequently evaluated at q_I=0. From Eq. 18, the second term in the right-hand-side is zero. From Eq. 24, the third term in the right-hand-side is -N_A[1-4πR_I³/(3 L³)]ξ'_S
• Page 7, Eq. 33: In contrast to the preceding one, this equation considers q_I to be the full ionic charge. Following from the previous comment, the second term in the parenthesis of the right-hand-side should be zero, i.e. the equation is incorrect and should read ΔG_C1^LS = 4πR_I³/(3 L³) (N_Aq^IΦ_o^raw+ΔG_C1^LS). Note that the error affects the article but not the calculation results, which actually relied on the correct equation

[86] Reif & Hünenberger, J. Chem. Phys. 134 144104 (2011)

• Figure 2: a strange "20p0" artifact (in black) has appeared in the middle of the figure upon printing (should be ignored)

[88] Horta et al., J. Chem. Theory Comput. 7 10161 (2011)

• Table 4, entry ether/OE: charge in 53A6_variant should be -0.420 instead of 0.420
• Table 4, entry aldehyde/O: charge in 53A6_OXY should be -0.475 instead of 0.475
• Table 7, entry 53A6_OXY for DXE: hydration free energy ΔG_wat should be -16.1 +/- 1.8 kJ/mol instead of -18.5 kJ/mol +/- 1.8 (the reported value was mistakenly calculated using a slightly modified OCCO torsional-energy term), see [98] Table 6 where this error was reported (this value is also recalculated therein using a slightly different calculation protocol but the normal OCCO torsional-energy term of 53A6_OXY, resulting in -14.7 +/- 0.1 kJ/mol)
• Page 1023, left column, line 23: a value of 4.575·10^-4 (kJ·mol^-1·nm^-3)^-1 was used for the simulations in water, but it is not the value appropriate for pure water, which would be 7.505·10^-4 (kJ·mol^-1·nm^-3)^-1, see corresponding note under [81]; the error incurred is probably negligible, considering that this parameter is combined with the arbitrary barostat coupling time, which has itself a very limited influence on the thermodynamic properties of the system

[90] Choutko et al., Chem. Phys. Chem. 12 3214 (2012)

• Page 3219, left column, lines 29-30: "a slight excess of W₂^L over W^L water units is observed at the wall" should be "a slight excess of W^L over W₂^L water units is observed at the wall"

[93] Kunz et al., J. Comput. Chem. 33 340 (2012)

• Page 342, Eq. 18: the first summation over index i should be deleted, and the second summation replaced by a summation over j≠i instead of j>i.

[98] Fuchs et al., J. Chem. Theory Comput. 8 3943 (2012)

• Table 7, entry rms for 53A6_OXYrec: "[0.03]" should read "[0.04]" (rounding error).
• Table 7, entry rms for 53A6_OXY+D: "[0.97]" should read "[0.03]".

[99] Laner et al., Mol. Simul. 39 563 (2013)

• Page 579, left column, line 11: in terms of entropy, the factor is about 2.7 (122.2/46) and not 3.8, see [Laner,KIN] p XXX where this error was reported (a value of 2.5 is actually proposed therein based on more extensive simulations)
• Page 576, Figure 4, points at n=11: the values for these points are affected by a small calculation error and are too low (this problem also affects [81], Figure 5), see [104] (caption of Figure 9) and [105] (caption of Figure 8) where the error was reported and correct values for these points can be found.

[101] Wang et al., Eur. Biophys. J. 42 521 (2013)

• Page 534, left column, line 40: "it contributes to affinity and not to specificity, and is thus a steering force" should read "it contributes to specificity and not to affinity, and is thus a steering force"

[102] Rocklin et al., J. Chem. Phys. 139 184103 (2013)

• Page 17, caption Table II, and p 23, caption Table IV : the value of γ_S should read 0.00764 e nm² (2*0.417*0.09572² for TIP3P water; instead of 0.0764 as written); this is just a typo, all calculations were performed with the correct value.

[110] Bieler et al., J. Chem. Theory. Comput. 11 2575 (2015)

• Table 1 : the numbers reported in this table are not the numbers of solvent molecules (as written in the caption), but the total numbers of atoms in the systems, mistakenly increased by one; the number of solvent molecules can easily be calculated from these as N_svt=(N_atm-A-C)/B where A is the number of solute (united-)atoms (12 for 12C4, 15 for 15C5, 18 for 18C6), B is the number of (united-)atoms per solvent molecule (3 for H₂O, 4 for DMSO, 3 for CH₃OH), and C=2 (one for the ion, one for the erroneous +1); see [128] (Table 1 and its caption), where the error was reported and corrected values can be found.
  

[114] Horta et al., J. Chem. Theory. Comput. 12 3825 (2016)

• Suppl. Mat. B table B.4: the simulation values of ΔH_vap (sim) and the deviations from experiment (exp) for the 2016H66 parameter set mistakenly excluded the contribution +RT (2.48 kJ·mol^-1 at 298.15 K) corresponding to the ideal-gas pressure-volume term; this mistake consistently affects the figures and RMSD/AVED reported in the main manuscript, but not the quality of the parametrization itself (which had been carried out against the correct values); the correct RMSD/AVED values for ΔH_vap would be 4.4 and 2.6 kJ/mol (calculated for 58 liquids, also including water from [124]; instead of the reported 3.5 and 0.2 kJ/mol for 57 liquids); see [124] (Table 1 and its caption), where the error was reported and corrected values can be found.
• Suppl. Mat. B table B.1: the CAS indexes given for 6 of the alcohols are incorrect; the correct CAS are 71-41-0 for PTL (ant not 111-27-3), 111-27-3 for HXL (and not 111-70-6), 111-70-6 for HPL (and not 111-87-5), 111-87-5 for OTL (and not 67-63-0), 67-63-0 for 2PL (and not 78-92-2) and 78-92-2 for 2BTL (and not 71-41-0)

[124] Gonçalves et al., J. Chem. Theory. Comput. 15 1806 (2019)

• Table 1: is affected by the same mistake as Suppl. Mat. B table B.1 of [114];  the CAS indexes given for 6 of the alcohols are incorrect (see above under [114])

[132] Kubincová et al., Phys. Chem. Chem. Phys. 22 26419 (2020)

• Page 26431, left column, line 8: "4-6 SW or 0.75 SH" should read "4-6 SH or 0.75 SW"

[137] Kubincová et al., J. Chem. Phys. 155 094107 (2021)

• Table 1: in the caption, "root-mean-square dispersion and repulsion coefficients" shoud read "square-root dispersion and repulsion coefficients"
• Reference 73 : is incorrect; should be "Lin & van Gunsteren, J. Chem. Theory Comput. 9 (2013) 1328"
  

[BOOK]

• see specific web page for the book erratum