MolBioWiz by JoeCouto

Several good molecular biology programs are available for free so, what is special about MolBioWiz?

MolBioWiz works on fastA files, so you can process thousands of sequences at once.
MolBioWiz works locally on your computer, using your browser, so, there is no waiting for servers to process your data and unless someone hacked into your computer your data stays in your computer. You can test this: after opening MolBioWiz in your browser, disconnect your computer from the internet. MolBioWiz will continue to work just fine.
MolBioWiz has some tools for text and sequence processing that are not found together in other programs.
MolBioWiz can align V antibody sequences with unusual FR1 residues, which occur in some rabbit V-regions. I will consider including other unusual Ab V-regions if anyone sends me such a request (joe.couto@abaeternum.com).
MolBioWiz has an Ab developability problem prediction tool. NOTE that the detection of such sequence motifs does not mean that they are actually problematic.

There are 3 text areas: Top Mid Bot.

Usually, you upload or paste text or sequences into Top.
Then, you use the MAIN FUNCTIONS to process your data.
In most cases the results appear in Mid.
Buttons above the 3 text areas Top Mid Bot
- clr - clear text area
- Mid/Bot/Top(replace) - copy to Mid/Top/Bot replacing previous contents
- Mid/Bot/Top(append) - copy to Mid/Top/Bot appending to the bottom of previous contents
- save as - download text area contents as a text file
- fastA⇄name::seq - works only within Mid. It toggles the sequences format between fastA and a single line where names are separated from sequences by :: NOTE that this only works if sequences are in single lines - use the seqs button to make it so. The reason for using :: instead of just : as a separator is that some seq names contain a colon. NOTE: you can use the text/replace function to replace : with :: or vice-versa.
- delete duplicates - works only within Bot. If you have many lines where some of them are duplicated, this button will eliminate the duplicates. It ignores leading or trailing spaces, but it is case-sensitive, e.g. A is not seen as a duplicate of a.
- show - see Antibody Developability in AbALIGN.

The UPLOAD button opens a dialog for selecting and uploading the contents of TEXT FILES into one of the three text areas.

You can try uploading other types of files, such as Microsoft Word files, but you will get extra "nonsense" characters.

If you want to work with text from web pages, pdf files, etc, you can highlight the text, copy it (ctrl + c), and paste (ctrl + v) it into these text areas.

grep - Enter a keyword in txt1 above and press grep. MolBioWiz will search for that keyword in Top and extract the lines that have it to Mid and the lines that don't have it to Bot. Check the case-sensitive box if you want the search to be case-sensitive. KEYWORDS can be REGEX (regular expressions); for example, the REGEX for N-glycosylation sites is: N[^P][ST]. To learn more about REGEX, google javascript regex.
findAll - This is like grep but the output is a list of found motifs and their positions. For example, the output from a search of N-glycosylation sites might be: 8:NHT 143:NGT 161:NNS.
replace - This like your typical find/replace except that it does not destroy the original text. Enter ABC into txt1 and a XYZ into txt2 and press replace. MolBioWiz copies Top to Mid and replaces all found ABCs with XYZ.
substring - This function extracts a column of text between the two indices. For example, if you had the three lines shown below to the left of the arrows in Top, and you entered 3 and 5 and pressed substring, the result in Mid would be the three lines shown to the right of the arrows. (⚠ note the index numbers in the first line, they are not intuitive, but that's what programmers do).
012345678 → 34
ABCDEFGHI → DE
JKLMNOPQR → MN
other text functions - At least with Chrome or Edge browsers you can press crlt + F to search text, and you can right-click over the text areas for additional text functions.

Seqs works on fastA-formatted sequences entered in Top. MolBioWiz can process thousands of sequences.

Examples of fastA-formatted sequences:
>seq1 name & metadata in a single line starting with >
ATG--TTTGgacctacctCGT---GGcc
TGAGATT---TAGGCGCTCAA---
>seq2 seqs can be broken and have numbers and dashes
EVQLVESGGG LVQPGGSLRL SCAASGFTFS SYAMSWVRQA PGKGLEWVSA 50
ISGSGGSTYY ADSVKGRFTI SRDNSKNTLY LQMNSLRAED TAVYYCAK 98
fastA⇄name::seq - this button is located above Mid and works only within Mid and when each sequence is in a single line. It toggles between fastA and a name::sequence single line format. A double colon :: is used as a separator because colons appear in some sequence names. NOTE that you can use the text/replace function to replace : with :: or vice-versa.
XYZ removes dashes, numbers, and other non-sequence characters, and outputs a single line of upper-case sequence to Mid.
X-Y removes blank spaces but keeps all other characters and the original case.
RC (for DNA only) outputs the reverse-complement in a single line of upper-case nts.
wordSize and lineSize respectively insert spaces and line breaks after n and m sequence characters.
show res #s radio button → select no or yes.
Accepted characters

If sequences have the non-IUPAC letters J or O, the output will contain a blank seq line following the >name line, and, a warning/error will appear in red (top right corner).
RC (rev-comp) will process only IUPAC codes as follows
A-C-G-T-U[AG][CT][GC][AT][GT][AC][CGT][AGT][ACT][ACG][ACGT]
A-C-G-T-U-R---Y---S---W---K---M---B----D----H----V----N
T-G-C-A-A-Y---R---S---W---M---K---V----H----D----B----N
>seq1
ACGTURYSWKMBDHVN
>seq1_RC
NBDHVKMWSRYAACGT

transl Translates nucleotide sequences (⚠ AGCTU nt only)

wordSize lineSize and show res #s - these are formatting options as described above.
transl from (5'or motif)

by default, the 5' radio button is checked and translation starts from the 1^st nt. You can force all sequences to start at nt #; for example, set at nt # = 8 → atacgccATGGTT translates to MV.
if the txt1 radio button is checked, MolBioWiz will search each sequence for the motif entered in txt1 and start translation from the stipulated nt #; for example, enter the Kozak motif C[AG]CCATGG in txt1 and set at nt # = 5 → both atacgccATGGTT and taggattatacaccATGGTT will translate to MV. ⚠ NOTE1 if you have nothing entered in txt1 the translation will be as if the 5' radio button were checked. ⚠ NOTE2 if the motif entered in txt1 is not found in one or more of the sequences, they will not be translated and a red error message will appear on top right.

Translation Frames Choose any combination of frames 1, 2, or 3.
DNA over trnsl? If you select radio buttons to show DNA above translation, the single-letter amino-acid residues will be centered below the respective codons. The chosen line lengths will correspond to the number of amino-acids per line. You cannot break the lines into words.

name This function allows you to change sequence names (it is useful if you have many sequences).

padding makes all the sequence names the same length as the longest name (by adding _).
replace replaces txt1 with txt2.
splice keeps only part of the names, according to the start and length parameters. It's OK if length is 0 or if it exceeds the number of leftover characters. In those cases the new name keeps all characters from start through the end of the name.

find This function extracts sequences that have a pattern entered in txt1, into Mid and sequences that don't have it into Bot.

fetchFromUniprot fetches fastA-formatted protein sequences from UNIPROT.

Enter a list of Uniprot accession numbers (e.g., Q9NZQ7 Q01279) in Top and press the button.
NOTE: the list can be in a column or all in a row or both but make sure there are ONLY BLANK SPACES in between the accession numbers.

Properties works with fastA-formatted protein sequences entered in Top.

The output is a table with Aa frequency, size, molecular mass, charge → at pH, the weighted arithmetic Kyte-Doolittle and Hopp-Woods means, name, and sequence.
It is best to copy and paste the output to a spreadsheet.
The charge → atpH pair of values is meant to give you the pI, that is, the calculated pI is the pH if the charge is close to zero. The algorithm is as follows:
1. calculate the charge of the protein at pH6.5 using the Bjellqvist's pKa values: R(12) K(10) H(5.98) Nterm(7.5) D(4.05) E(4.45) Y(10) C(9) Cterm(3.55)
  This is done by adding the following two sums:
  for positive Aas and Nterm: ∑ 1/(1+10^pH-pK_Aa)
  for negative Aas and Cterm: ∑ -1/(1+10^pK_Aa-pH)
2. if the charge is negative lower the pH by 1 unit; otherwise, increase the pH by 1 unit
3. re-calculate the charge
4. iterate steps 2 and 3 but if the charge changes from pos to neg or vice-versa add or subtract only half a pH unit.
5. keep iterating and halving the pH units until the charge is within 0.01 of zero, or until the number of iterations reaches 30.

AbAlign This function works on fastA-formatted antibody variable region sequences entered in Top.
Developability Buttons These buttons work only on Ab V-regions that are aligned exactly as the AbAlign output.

The sequence alignment outputs to Mid and shows only the variable region.
⚠ Sequence names are truncated to 10 characters.
A second row of buttons appears between Mid and Bot see Antibody Developability below.
The tool uses the IMGT FR numbering system.
The CDRs are equally filled in from the N and the C-termini. CDR3 can accommodate up to 25 residues.
There are five extra residues in FR1 to accommodate unusual rabbit sequences. One is at position 0 and four are between positions 11 and 12.
The tool finds VH, VK or VL frameworks using a small database of template sequences.
One can only change these templates at the code level. You can, however, change the percent ID that the tool uses to find a match. Hint: if you know that a FR is NOT present in your sequences set that FR parameter to 100% to prevent the tool from finding a similar but wrong FR sequence.
If the tool can't find a complete FR, it tries to find a truncated one using the last two parameters, minimum length of truncated FR and %ID, respectively.

Antibody Developability

The developability buttons appear automatically when the AbAlign button is pressed. Alternatively, you can press the show button.
Use the developability buttons to find sequence motifs that might cause Ab developability problems
If found, the motifs show up in color and bolded
note that the sequences are formatted slightly differently in that the initial greater-than sign and all the dashes are replaced by underscores
glyc - N-linked glycosylation sites N[^P][ST]
NxDx - CDR motifs that may lead to deamination/isomerization N[GSTNH]|GN[FY]|D[GSDTH]
specf - CDR3 scFV non-specificity motifs GG|RR|VG|VV|WW|YY|W\wW
aggr - motifs in CDRs that cause aggregation FHW
visc - motifs in CDRs that lead to viscosity HYF|HWH
posit - marks R,K,D,E CDR residues if Nmbr(R,K) - Nmbr(D,E) > 1
cys - colors canonical Cys residues at positions 23 and 104 in red and others in blue

pepScan scans fastA-formatted protein sequences for good peptide immunogens.
The program outputs:

a list of selected peptides in Mid
aligned peptides, relative to each protein sequence, in Bot

Peptides are displayed with their respective scores - higher scores are better.
Each peptide gets an initial score = peptide length + number of unique residues.
Points are added for good solubility (see maxKyDo).
Points are subtracted for:

localized hydrophobicity (see ILVMFAC)
localized charges (see EDRK)
lack of sequence complexity (see duplc)

How to best look at the aligned output

Download the output and use a text editor with text wrap turned off. WordPad works great.
Alternatively, split the Bot text area, place the sequence on the top and scroll to see the aligned peptides below. The scroll bars are not synchronized but it is easy to sync them visually.

Selection Parameters

minPepLen Exclude peptides that are shorter than the entered number.
maxPepLen Exclude peptides that are longer than the entered number.
maxKyDo Exclude peptides with a KyDo higher than the entered value. KyDo is the weighted arithmetic mean of the residue Kyte-Doolittle values. The NH3+ and COO- termini are not considered. Negative KyDo values are indicative of good peptide solubility and possibly also good immunogenicity. Score Adjustment: 2 points are added if kydo is between -0.2 and -1; 1 point is added if kydo is less than -1 (a KyDo that is less that -1 is probably due to the presence of many charged Aa's which may lead to isolating mAbs with lower specificity).
EDRK ILVMFAC duplc It is best to leave these parameters alone.

duplc If any Aa is present more than twice the peptide score gets decreased by s = 2^p(n-2), where n is the number of repeats and p is the entered parameter. (This is my own formula - feel free to contact me about it: joe.couto@abaeternum.com).
EDRK The score is decreased as described above if more than two charged Aas (Glu Asp Arg Lys) are present in a row.
ILVMFAC The score is described above if more than two hydrophobic Aas (Ile Leu Val Met Phe Ala Cys) are present in a row.

forbid The default is BJOUXZ. If any of these letters is present in the peptide, it will be discarded. You may, for example, add W to the list if you want to exclude peptides with tryptophan.
score and bestN radio buttons: you can either choose a minimum score threshold below which peptides will be discarded or, choose to display the peptides with the best n scores.
nextStart The next potential peptide be considered starts n residues after the previous peptide.